Analytics

June 3, 2009

Data Visualization Dresses Up Search - SEM Synergy Extras

Recent developments, like Microsoft Bing, Google Search Options and Yahoo's Web of objects strategy, are moving search engines toward more sophisticated models of data visualization. As new, useful ways to present information to the general public are developed, search marketers are reminded that improved data visualization models could benefit the community in many ways.

Photo by Jesslee Cuizon via Creative Commons

During today's weekly podcast of SEM Synergy, we explored data visualization applications in the search industry. My guest was Richard Zwicky, founder and CEO of Enquisite. Enquisite provides analytics solutions that visually organize and structure data, with paid and organic search professionals in mind. Involved in Internet marketing tracking solutions for almost a decade, Richard has seen an evolution in the display capabilities of analytics solutions, as well as in the expectation of search marketers regarding those display capabilities.

But data analysis is just one of several applications that could be improved through enhanced visualization. Data analysis, data reporting and visual search engines are three platforms in the search industry that will undoubtedly see changes within data presentation in the future.

Visual Data Analysis

Eye-tracking studies, long-tail graphs, map overlays -- data can be gathered, parsed and presented in a myriad of visually enhanced ways. The exciting thing about analyzing data visually is how the right format can make hidden trends and findings readily apparent. Knowing that X number of visitors came to your site in a week is great. But seeing the number of visitors to your site over a year represented in a timeline graph? The latter is clearly more valuable. Of course, that same data would be less suited for something like a pie chart or word cluster. The right visual application can relieve an Internet marketer from having to sift through mountains of data in order to locate the important stats and trends.

Visual Data Reporting

As a relatively new profession, Internet marketers can find themselves evangelizing to their clients or company executives about the opportunity of search. With data in hand, the potential of search marketing ROI can be very convincing. Proof is what company execs are looking for, and data makes a great argument. But overzealous search evangelizers should steer clear of overloading listeners with excessive data. It's important to only present the most useful data in an easy-to-understand visual format. If the person watching the presentation has to decipher how to read the data in the first place, there's a problem. But if the person watching immediately understands the data by simply looking at the projector, the point has been made.

Search Visualization

Along with data analysis and reporting, data visualization has the potential to aid in the understanding of many topics and concepts. Data visualization uses of the general public are being addressed by a number of visual search engines, like Viewzi and Searchme. About a year ago, Danny Sullivan saw visual search as little more than "eye candy", explaining that the shortcomings of traditional search were not being addressed by available visual search products; in fact, most of them introduced new problems all their own. Where we're likely to see real progress in visual search is within the mainstream engines. Google Squared could potentially change searcher behavior and Yahoo's updated image search demonstrates an improvement on the old standard.

The possibilities for data visualization are endless, and as technology advances, we're sure to see a growth spurt in the industry. Enquisite is certainly one of the innovators of data visualization for Web analytics, and with a new platform, Enquisite Campaign, now available, do yourself a favor and check them out. Thanks to Richard Zwicky for coming on the show to explain how visualization of data can help with analysis and reporting amongst search practitioners. To find out more from Richard, you can read the Enquisite blog or attend his weekly Ask the Expert round table discussion beginning June 16.

Posted by Virginia Nussey on 06/ 3/09 at 3:37 PM | Comments (0)
See more entries in Analytics, SEM Industry, SEM Synergy, Search Engines

February 16, 2009

Six Questions with John Marshall

SES London will be kicking off tomorrow. Last month we brought you interviews with Greg Jarboe and Debra Mastaler, two Internet marketers who will be speaking during SES London.

To close out our series of SES London Q&As, John Marshall, CTO of Market Motive, answered six quick questions. Or six questions quickly. Either way, enjoy!

1. You'll be speaking on the panel Measuring Success in a 2.0 World. What "cutting-edge" statistics should site owners be paying attention to in a 2.0 world? Have the metrics changed in the new Web environment?

I think the problem is that these cutting edge statistics are getting blunted by the 2.0 world. Increasingly your content exists on websites other than your own, so it's hard to measure.

2. Search Resources: Building A Better Economic Model is another session you'll be speaking at. According to the description, part of the premise is that only the measureable marketing activities that prove their worth will survive the chopping block. Have you found that paid and organic search marketing operate with different levels of success or is it like comparing apples to oranges?

You need them both, and neither is headed for the chopping block!

3. Another session you're speaking on is Landing Page Testing & Tuning. Can you give us a rundown of some of the landing page testing methods that people should have in their arsenal?

With free tools like Google Website Optimizer, the important thing is that you continuously test.

4. You were interviewed here at the Bruce Clay, Inc. blog last May by Lisa Barone. She asked you about the most important metrics to pay attention to, as well as the most overrated. In your opinion, has anything changed since then? Is it still about average time on site (ATOS) and not as much about ROI?

Yes, I'm still a big fan ATOS. ROI is just too complex and prone to problems.

5. What would you say to someone who feels that analytics data is so comprehensive that it's hard to take action? Are there any reporting solutions available to help this struggling SEO narrow down the actionable items?

If you can't make sense of ATOS (or its cousin, bounce rate) then you should ignore the web analytics data and survey your customers.

6. While you're at the conference, what must-see sessions do you plan to attend? Where will other attendees be able to track you down?

I'm really looking forward "Why Does Search Get All the Credit" on Tuesday.

I think John was intentionally coy with his answers so that you, dear readers, would be compelled to attend SES London to find out more. You can also check out his past interview here on the BC blog when John spoke to Lisa in preview of last year's eMetrics San Francisco. Thanks for stopping by the blog again, John!

Posted by Virginia Nussey on 02/16/09 at 2:03 PM | Comments (0)
See more entries in Analytics, Interview, SES London 2009

February 11, 2009

Analyzing & Converting SEO Traffic

I think I have a sickness. I can't stay away from the analytics sessions. This time, my addiction has lead me to this great panel with Craig Hordlow, Red Bricks Media, Jim Sterne, Target Marketing, Lauren Vaccarello, lv logic, and Richard Zwicky, Enquisite, who was a recent guest on SEMSynergy. Everyone's favorite Danny Sullivan, Search Engine Land is going to moderate.

Jim Sterne is going to start us off because he's totally the analytic guru. He writes about the theory of search and analytics. He's going to talk about what's on your mind: SEO, SEM, social media, behavioral targeting, 301 redirects, image search, branded search... so many many things. But to the higher ups it's just Search. And they have a lot on their minds too and all they really care about is their bonus. The nice thing is that everything you do in search affects their bonus. In order to explain it to them, don't try to give them every piece of information. Protect them from that and instead give them a promise. "If you give me the tools, money and people. I will increase sales/contacts/etc. by X percentage in Y months".

When they say "how are you going to do this?" you will be tempted to explain. Don't do it. Stick to the ROI. If they insist that they want to know? Send them to a conference like SMX West where smart people will explain it all to them.

Hee. I love Jim.

Lauren Vaccarello is next and I love her too. She's making a BSG joke and that's why she's awesome. Her disclaimer: she's using Google analytics but these recommendations can be applied anywhere.

Finding your Larry Fitzgerald: [translation: your top performers]

Set up a goal and find the keywords that make you money. Find the high performance pages with high conversion rates and low bounce rates. It's not about visits, it's about conversions. Just bringing people in isn't enough.

Finding your Jackie Moon: [translation: your worst performers]

Create bounce and transaction reports to find the pages that don't convert or hold attention. You do this by comparison. If all your pages have a 30 percent bounce rate but then you have one page with a 50 percent bounce rate, that's a Jackie Moon. However, you need to actually look at the keywords bringing people to that page: figure out which keywords cause people to bounce.

The Felix Gaeta pages: [translations: pages with potential that just fail] [I'm just going to be over here giggling to death.] [I hope you're not the only one that understands what's going on. --Virginia]

You need to understand behavior. Find the pages with very high page views per visit. These aren't the grazers who just come back, steal a picture and then wander away. These are the ones who hang around but don't convert. The problem is you. Someone doesn't go through 21 pages without buying something unless you won't let them. Look at the pages and figure out what's standing in their way. Figure out if you're not matching expectations.

Branded versus non-branded terms

Your company, yourcompany.com -- branded
Sports magazine -- non-branded

Honesty is important. You need to separate your branded terms from your non-branded terms. Why separate it out? Because branded terms convert better, they don't count. You want to increase conversions on the non-branded terms.

Caig Hordlow follows Lauren.

There are two fundamental human interactions: connections and disconnections.

Are you really listening to the whole conversation? Or are you just focused on the "top" pages, keywords and numbers. You have to listen to the whole conversation and dig deep.

Motive analysis:

• Identify the motive of visitors by understanding the nuances of the search query.
• Segment the queries by motive.
• Analyze the performance/entry pages. Tip: generate a list of filters and segment.
• Identify motive disconnects.
• Modify entry pages to solve problem.
• Grow your site intelligently.

Basic necessities: tools and a brain

Ten motives of search:

1. Product/service
2. Comparison/quality
3. Adjective qualifier
4. Intended use
5. Vendor/manufacturer
6. Location
7. Action request
8. Instruction
9. Definition
10. Problem

He likes to use personas. Specifically his mother because she's the lowest common denominator. Aw.

Example: Shoe shopping

Three personas:

• Shopping addict: "Gucci", "tobacco pumps"
• Local enthusiast: "Toronto", "hiking"
• Bargain hunter: "cheap", "sales"

What are my top motives? Which motives are underperforming? Why are my underperforming pages not connecting with my visitor? You'll need to dig in to find micro-segments. That's easier to do with PPC than SEO.

So what do we do about it? Bring it back to the entry page. Review each page for motive disconnects. Is the message that the user wants to find on the page? Is it succinct and persuasive with more info easily available? Is it even feasible for you to be relevant? Maybe you're not cheap or discount. Is it visible enough? Should the message be on its own page? You don't want to clutter your pages to the point that there is a lack of focus.

You're either going to rebuild your entry pages or build new entry pages.

Microsegmentation grows your long-tail intelligently. You'll need to build out new pages and a lot of them.

Richard Zwicky from Enquisite is last.

There's a difference between Web analytics and Search analytics. It's about specialization. He's going to be going through a lot of case studies. I'll focus on the takeaways.

Page-two terms are easy wins. You're nearly there. You're already considered valuable.

Don't skip the small stuff. Think of it like a score out of 100. If you don't do one, your best score is 99. If you skip two, your best score is 98. Don't give up those points without trying.

To determine the potential of a keyword use the following formula:

Keyword potential = (1 minus keyword volume / total search engine referrals for the site)(Average page views / visits for that phrase)(time on site in seconds)(1 minus bounce rate)

Then this has to be normalized against all the other terms on your site. This can also be run for actions and conversions. It's built into the Enquisite tools.

Understand how a searcher's physical location impacts your search results and business. Personalization means that you might not rank well if you don't have regional links into your site.

Q&A

How do you reduce bounce rate?

Jim: Look at what's bringing them there. You might be drawing people for the wrong purpose.

Richard: Take a look at the page and see what's bringing them in the page. Also, deoptimize for non-useful words.

Lauren: Consider if there's another page on your site that should be optimized for that keyword instead.

How do you test and optimize WordPress?

Lauren: You can test the content on the page. Add more to that page that might help support it.

Richard: The content management system shouldn't matter.

Craig: Figure out what you're testing first. Most people have no idea.

In terms of visibility scoring/ranking reports, I thought there would be more discussion about it. Is it important?

Danny: They talked about it in the previous session. Showing up doesn't necessarily mean you're doing well, except in terms of reputation management.

Lauren: It really only matters if you're on page two and you're getting conversions there. Moving to page one is a huge opportunity.

Richard: On average page two to page one is a 45 percent increase in traffic.

Danny: The value is not a ranking report. It's an opportunity report where you correlate conversions to the ranking.

How do you control the snippet that shows up in the SERPs?

Craig: The Meta Description.

Lauren: Match the page description to the keywords bringing you traffic.

Danny: Usually they use the Meta Description. If the query didn't show up in the Meta Description, then they'll try to pull from the text on the page.

Did you say that the search algorithms will value the business address in terms of linking?

Richard: Yes, they'll try to determine the location. So that they can say someone in this area finds this site that they're linking to useful. So if you're in Houston and you link to Enquisite, the engines will say people in Houston consider this site useful. It's not as important for national businesses but it's incredibly important for local businesses.

Posted by Susan Esparza on 02/11/09 at 4:25 PM | Comments (0)
See more entries in Analytics, Liveblog, SMX West 2009

February 4, 2009

New Wave Web Analytics - SEM Synergy Extras

It's that time again! SEM Synergy, baby!

On today's show I sit down with Richard Zwicky, CEO of Enquisite, to pick his brain about Enquisite's search analytics platform, analytics in the Web 2.0 space, and how a payment model based on conversion value could change the SEO industry. The most fascinating part for me was when Richard explained that a performance-based Internet marketing economic model is just around the corner thanks to advancements in assigning values to conversions. From our interview:

"We're on the cusp of actually being able to measure everything in the same way and value things the same way as we traditionally have in PPC, [...] on actually tying it to that conversion -- whatever conversion means to you. [...]

It doesn't really matter what your goal is of conversion, of value point. What does matter is that if you can tie a value to that conversion and then a value to the customer acquisition, you can pay your SEO firm, your email marketing firm, the firm you use to do newletters or email blasts -- it doesn't really matter -- you can pay everybody for performance. Do a good job, bring me lots of customers and as I make a sale we both win, so there's no maximum cap."

If you too are intrigued by how a technology like the one Richard describes could change the SEO game, listen to the episode over at SEMSynergy.com or download it from WebmasterRadio's SEM Synergy archives.

While we're on the subject of the new wave of analytics, yesterday Ralph Wilson posted an interview he had with John Marshall, founder and CTO of Market Motive during PubCon. A Web analytics expert, John talks to Ralph about information you can gather that analytics data is not able to tell you.

The needle on my cool-o-meter went off the charts when John mentioned a tool he recently discovered called ClickTale. A long-standing opponent of site path as a viable search marketing metric, John changed his tune after playing around with ClickTale. The product lets site owners actually watch every move a user made on the site -- each movement, hesitation and click of the mouse recorded in a movie. After watching a few of those I bet you'd have a pretty good idea of some of the hang-ups and abandonment points on the site.

With SMX West and SES New York barreling down on us, I'm sure this isn't the final word on improvements to data gathering, analysis and reporting. I'm looking forward to those in the coming months, but if I missed any notable advancements in the world of analytics or things you'd like to see in analytics platforms that would make them more useful, let me know in the comments, would you?

Thanks again to Richard Zwicky for coming on the show. Richard will be attending SES London next week and is speaking during several sessions. If you're in town I'm sure he'd love to answer your analytics questions. Then, when you're ready for an analytics platform designed just for search marketers like you, remember that Bruce Clay, Inc. will soon be offering Enquisite Pro and all the smarty-pants data that comes with it to clients.

Posted by Virginia Nussey on 02/ 4/09 at 5:08 PM | Comments (0)
See more entries in Analytics, SEM Synergy

January 7, 2009

SEM Synergy Extras

Unless Thanksgiving put you in a food-induced coma for the last few months, you'll have noticed that the holiday season has come and gone. Considering the delicate economic climate, those in the online marketing space were watching carefully to see how online consumer spending fared during what is typically the most productive time of year for retailers.

On SEM Synergy today the subject du jour is online marketing and SEO considerations for ecommerce sites. Of course it's too late to take action on your ecommerce site in order to improve its effectiveness for holiday sales, but it's never too late to improve your retail site going forward.

Now that Black Friday, Cyber Monday and Green Monday (I hadn't heard of the latter before, but apparently Green Monday is the last day to buy products online and have them delivered before Christmas without upgraded shipping) [Also known as the day I do all my shopping. --Susan] are behind us, stats are coming out about online spending in November and December. From the looks of it, online sales were disappointing for all but a few big name retailers like Amazon and Walmart.

According to Compete.com, the 10 leading online retailers saw year-over-year growth this holiday season:

Unfortunately for the little guys, they didn't see the same strong numbers. At this point you can see why it is so important for small and midsize companies to implement top-notch online marketing campaigns in order to keep up with the retail giants. But how?

Content and Design

One important aspect that can almost always be improved is site content. According to an online customer experience survey, more than two-thirds of users want to see imagery and details in product descriptions. It's good to note that catalog descriptions are typically not the best product descriptions to use for SEO purposes. A catalog description often does not contain the keywords that searchers are using to find that product. Write content that creates a sensory experience using descriptive adjectives and words that the typical searcher would use to find your product.

Related to text content is the visual content of the site, or the design. Include products on the home page, use images above the fold, prominently display a clear value proposition and don't clutter the page.

The CMS

A frequently encountered roadblock to retail site SEO is a poor content management system. While a CMS can be a great tool for managing a site with a lot of content, you want to be sure that the system is search friendly. A search-friendly CMS offers flexibility and freedom, such as the ability to customize page Titles, Meta tags, and URLs. When it comes to URLs, you want to make sure that there are as few hindrances as possible for search engine spiders crawling your site. URLs with session IDs and multiple dynamic variables can be an obstacle to spiders.

Shopping Cart

Each extra step in your shopping cart process is an extra opportunity for your customer to bail on the purchase. If possible, distill the checkout process to two pages and cut out any unnecessary fields. If I'm ready to give you my payment info, maybe it'd be best to let my buy without registering for an account first. Then, once I'm a happy customer I'll usually go back and create an account and fill out those extra fields you may have wanted but maybe weren't essential to getting that conversion.

Analytics and Segmentation

One of the most awesome things about setting up shop online is that everything is so measureable! Too often people forget to consider the data that they have available to them through Web analytics. Don't do that! Google Analytics lets business owners evaluate the traffic on their site to find and target the converting audience. Updates to the program have made it simpler than ever to segment the data, which makes the data more actionable. Separate segments like new and returning users, visitors coming from paid and organic, branded and non-branded search terms, etc. This way you'll have the easiest time finding actionable insights.

The Buying Cycle

Other than being there when they're looking for you, one of the best ways to prepare for a sale is being there when they're not looking for you. Become a familiar face and when the consumer is ready to buy, they'll come to you. The concept of being in all stages of the consumer's buying cycle is called halo media.

The idea is that by creating a circle of presence around your company through participation in multiple media channels both online and offline, your target audience will be aware of your product or service and, in turn, be more likely to find you if they need that product or service. Be in all phases of the buying cycle -- identifying a need, searching for a solution, finding sources, comparing sources, making a decision, affirming their choice, and establishing a relationship -- to increase the likelihood that you're who they turn to when they're ready to click "buy".

I want to thank today's guest Gregg Banse, one of the moderators of Webmaster World's ecommerce forum, for coming on the show. At Gregg's Web strategies and development company, 7th Pixel, he helps companies find strong footing on the sometimes unpredictable Web and he had some great recommendations for ecommerce sites looking to improve their online reach and visibility. To listen to the show, check out SEM Synergy on WebmasterRadio.fm or visit SEMSynergy.com.

Posted by Virginia Nussey on 01/ 7/09 at 3:16 PM | Comments (0)
See more entries in Analytics, E-commerce, SEM Synergy, Search Engine Optimization

December 30, 2008

Best of Search Conferences 2008: Day 3

Welcome to the final day of the Best of Search Conferences 2008! How's everyone feeling? It is New Year's Eve, ya know. And you totally deserve tonight's champagne after all the hard learnin' we just put you through. If you skipped out on the first two days, well, it's not too late to catch up.

Like any good conference host, I'm going to start off the day with a bit of housekeeping. At Bruce Clay, Inc. we're celebrating the last day of 2008 not only with the finale of our Best of Search Con but also with a special episode of SEM Synergy. Tune in today to WebmasterRadio.fm at 3 p.m. Eastern to hear guest Barry Schwartz give listeners an industry insider's take on 2008. Barry has been reporting on the industry for, like, forever in search years and offers an insightful look at trends and developments in the last year.

Also on SEM Synergy, Bruce gives us a preview of his predictions for the coming year (a taste of which you saw at PubCon), with more details coming in the January issue of the SEO Newsletter. And finally, Bruce Clay, Inc. is proud to announce that in 2009 we will be offering three SEOToolSet Training courses in New York!

Alright, thanks for sticking with me. Now, on with the show!

Keynotes

The Coca-Cola Marketing Metrics Journey, Part 2 - eMetrics Summit San Francisco, May 4-7
Speaker: Tim Goudie

• Even the world's biggest brands, like Coke, see ups and downs in the online marketplace. The important thing is to keep learning and evolving as you go.
• Data can be political, but giving everyone access to the data creates clarity and transparency in an organization. As good of a tool data is, it means nothing without the right analysis.
• Coca-Cola created a framework to continually measure brand health, brand advocacy, volume, media value, and marketing productivity. Make sure your metrics get at the fundamental objective of the business.

Keynote - Click: What Search Activity Tells Us About Society - SMX East, Oct. 6-8
Speaker: Bill Tancer

• Search queries are a window into human interests and can tell you more about an individual than they might tell you themselves. For instance, the fears most reported by those polled do not match the fears most searched for in a sample of 25 million Internet users.
• Intent is a factor that needs to be considered when making predictions based on searcher behavior. Tancer predicted third-place finalist Stacey Keibler would win Dancing with the Stars based on spiking searchers for the model's name. The problem was that searchers looking for images of Keibler were likely not voting for the contestant on the show.
• Follow the early adopters -- Young Digerati, Money & Brains and the Bohemian Mix -- for emerging trends.

All About Analytics

Top Takeaways:

• The process of analytics is to gather a report, analyze the data, optimize the content, and measure the change. You must have clearly defined goals beforehand otherwise measurements are useless.
• Create a document like Lippay's "grid". Include keywords, number of searches, conversion metrics ($ per PV, LTV, etc.), PPC data, paid inclusion (PI), algo, and search engine CTR by position. With this you can balance SEO, PPC and PI, find SEO referral gaps, find SEO content opportunities, and make traffic and value projections.
• On-site search can act as a productive data point. Searchers actively engaged in your site offer more data which leads to more opportunities for success.
• Segmentation is an essential part of analyzing your traffic. People aren't moving through your site with the same motive and you should be able to track that.
• The 3 Cs of Analytics are context, comparison, and contrast. Key main performance indicators (KPIs) are the time spent on the site, pages viewed, conversion rate, and cost and revenue of the visitor.
• Act on analytics data by improving keyword list efficiency and breaking out the best terms for SEO vs. PPC.

Analytics Every SEO Needs to Know - SMX Advanced, June 2-3
Moderator: Rand Fishkin; Panelists: Brian Klais, Laura Lippay, Jonah Stein, and Richard Zwicky

On-Site Search as a Crystal Ball - eMetrics Summit San Francisco, May 4-7
Panelists: Daniel Shields and Phil Gibson

Identify, Analyze, Act - SES San Jose, Aug. 18-21
Moderator: Chris Boggs; Panelists: Craig Danuloff, Brian Cosgrove, Heather Dougherty, Michael Stebbins, and Brett Crosby

Actionable Organic Search Analytics - eMetrics San Francisco, May 4-7
Moderator: Mike; Panelists: Matt Bailey and Diane Hoag

Legal Considerations

Top Takeaways:

• Trademark policies in the U.S. are different for each search engine. Google allows bidding on trademarks but does not allow the trademark to be used in the ad copy. Outside of the U.S. Google does not allow advertisers to bid on trademarks.
• The Communications Decency Act says that online authors and not publishers are liable for whatever they write. The law was meant to encourage publication and keep things from turning too defamatory online.
• Google is nearing a monopoly on search advertising market share. A monopoly is not illegal; the problem arises when a company tries to exploit its power to discourage competition.
• Have a contract with your consultants and clients specifying who owns what. Protect your property with management options if your consultant has access to any of your accounts.
• Behaviorally targeted ads raise the question of data ownership. The data is owned by the pary the consumer knowingly and willingly gave the data. Sharing of data requires opt-in consumer permission.

Legally Speaking: Recent Legal News About Search (February) - SMX West, Feb. 26-28
Moderator: Jeffrey Rohrs; Panelists: Clarke Walton, Sarah Bird, and Eric Goldman

Trademark Issues: What SEMs Should Know - SES San Jose, Aug. 18-21
Moderator: Jeffrey Rohrs; Panelists: Mark Rosenberg, April Wurster, and Eric Goldman

Googleopoly - SMX East, Oct. 6-8
Moderator: Jeffrey Rohrs; Panelists: James Grimmelmann, Shelly Palmer, Kevin Ryan, and Jimmy Wales

Legally Speaking: Recent Legal News About Search (October) - SMX East, Oct. 6-8
Moderator: Jeffrey Rohrs; Panelists: Mark Rosenberg, Jonathan Hochman, Roy Shkedi, and Deborah Wilcox

Advanced SEO

Top Takeaways:

• Link building tools include Link Harvester, Hub Finder, the Langreiter Tool, Search Status Tool, Google Alerts, and the Utility Linking Tool.
• The next generation link building strategies are link bait, article writing and content targeting.
• Mine your existing backlinks for opportunities to improve the anchor text. Start building a relationship with the site manager/owner.
• While it is irresponsible to use black hat SEO techniques on clients, the ethical nature of techniques often comes down to intent (for example, cloaking).
• Paid links aren't evil -- they're advertising. The Web is made for commerce. The line is drawn when the intent of buying links is to increase ranking factors.

Give It Up - SMX Advanced, June 3-4
Moderator: Danny Sullivan; Panelists: Rand Fishkin, Todd Friesen, Michael Gray, Rob Kerry, Marty Weintraub, and Stephan Spencer

Link Building Fundamentals - SMX East, Oct. 6-8
Moderator: Detlev Johnson; Panelists: Debra Mastaler and Eric Ward

Blow Your Mind: Link Building Techniques - SMX Advanced, June 3-4
Moderator: Greg Boser; Panelists: Jay Young, Stephan Spencer, Roger Montti, and Todd Friesen

Black Hat, White Hat: Playing Dirty with SEO - SES San Jose, Aug. 18-21
Moderator: Matt Bailey; Panelists: Bruce Clay, David Naylor, Jill Whalen, Greg Boser, and Todd Friesen

Advanced PPC

Top Takeaways:

• If you plan to target long tail terms, first bid on the most relevant keywords to build relevancy and then add less-relevant terms later on. This helps to establish a higher Quality Score.
• A granular search campaign will show the most targeted, relevant ads and receive a higher Quality Score. Be sure to send users to the most relevant landing page.
• When optimizing landing pages, remove banner ads, entry pop-ups and cluttered design.
• Use a variety of sources to create your keyword list. Use tools, the Web site, print material and press releases. But when researching potential keywords internally, be cautious of jargon.
• Segment your audience through the ads they click on. Think about who they are, how they reached you, and the location they come from.

Amazing New PPC Tactics - SMX Advanced, June 3-4
Moderator: Matt Van Wagner; Panelists: Addie Conner, Stanislas Di Vittorio, Siddarth Shah, David Szetela, and Natala Menezes

Landing Page Testing and Tuning - SES New York, Mar. 17-20
Moderator: Sage Lewis; Panelist: Tim Ash

Advanced Keyword Research Techniques - SMX East, Oct. 6-8
Moderator: Gordon Hotchkiss; Panelists: Christine Churchill, Ariel Bardin, and Marty Weintraub

Search 4.0: Search Ads and Behavioral Targeting - SMX West, Feb. 26-28
Moderator: Chris Sherman; Panelists: Kelly Gillease, David Kopp, Jonathan Mendez, and Natala Menezes

Happy 2009 everybody!

Posted by Virginia Nussey on 12/30/08 at 11:42 PM | Comments (0)
See more entries in Analytics, Best Of Search Cons 2008, Liveblog, Pay Per Click / Online Ads, SEM Events, Search Engine Optimization

October 28, 2008

Proactive Analytics Suite Launches

NuConomy announced the public launch of its Studio analytics product today. Previously in private beta, the free tools move toward what the company calls a "proactive" approach to Web analytics. Instead of reporting page views, a common metric, NuConomy Studio aims to present meaningful data in a way that borders on story telling. Check this out:

One of Bruce's favorite sayings is that you know half your marketing dollars are being wasted, you just don't know which half. Proactive analytics aims to identify that half. Rather than combing through and cutting up the data yourself, the tools compare that day's statistics with the historical record in order to report what it considers are "things you should know". Then, by simply logging on to the system, your email or your RSS reader, the reports arrive right on your screen. This seems to me like wonderful solution to the headache many companies face when they try to take on Web analytics -- a slow and labor-intensive process that begins with collection and then analysis, before conclusions can be drawn and action taken.

There are some sweet benefits to bloggers, too, who will be thrilled to see user activities, user interests and user contributions all mapped out pretty like:

Can you think of an easier way to address the most popular topics and encourage reader participation? Those posting videos will also get a kick out of the analytics reports that break down how much of a video is watched, who's watching entire videos and if ads were watched in the video.

Even more unique is the fact that NuConomy Studio is equipped to take action on your site. A two-way API effects automatic changes to a site based on the current metrics and correlations. The example they give is showing ads or pushing specific content that is relevant to a user's interests. Very clever, NuConomy, very clever.

But before we get too excited, we'll have to get a first-hand look at the quality of the NuConomy reports. If it really does pull out trends that you may not have spotted yourself or do much of the hard work for you, then these tools will see a boom in no time. But if the tool is simply pulling out all the weird spikes or outlying data points, you may end up with more noise than when you started. [I'll be Polly Paranoid and ask about security while we're at it. Who protects my data? --Susan]

Sure you had a 1500 percent jump in traffic from the UK, but if you don't sell anything in the UK, it doesn't really matter. The goal of analytics is to come up with actionable results from data that matters to you. Unfortunately, you can't throw a mountain into a sieve and hope that eventually you'll find the gold mine.

Have you tried NuConomy Studio? What do you think?

Posted by Virginia Nussey on 10/28/08 at 5:20 PM | Comments (0)
See more entries in Analytics, SEO Tools

October 6, 2008

Keynote - Click: What Search Activity Tells Us About Society

Danny Sullivan introduces the first keynote speaker of the conference, Bill Tancer, General Manager of Global Research at Hitwise, a Time Magazine columnist, and author of Click: What Search Activity Tells Us About Society. He says that we spend a lot of time wondering about who to market to in search. But oddly enough, for these machines, these search engines that have changed so much of our lives, there's relatively little information on how people interact with search. Bill's book tells us the behavior of people and shows us what they do.

Bill Tancer says that at his very first speech (which was for SES) he didn't know yet what he was going to say. It was standing room only and it was a data heavy presentation. He started off by saying that he loves data. In science camp as a kid he recited Pi to 200 digits. That's how much he loves data.

"To become aware of the possibility of the search is to be onto something." --Walker Percy

In Percy's book, he says that the most interesting thing about going into town is not getting there but the search involved in getting there. Bill's found something he likes even more than data, and that's the idea behind the click. Think outside the box. Think of what you can do with the data beyond buying search terms. Start thinking of all that's possible with search term data.

The data he's going to be using is based on the largest worldwide sample of Internet users (25 million). He and his team once did a test. They compared the top fears reported by people who were asked to the fears most searched for.

Reported fears:

1. Bugs, mice, snakes
2. Heights
3. Water
4. Public transport
5. Storms
6. Closed spaces
7. Tunnels and bridges
8. Crowds
9. Speaking in public

Most searched fears:

1. Flying
2. Heights
3. Clowns
4. Intimacy
5. Death
6. Rejection
7. People
8. Snakes
9. Success
10. Driving

He saw this as a great insight into what people are willing to claim in traditional market research versus how they really feel.

The Economy, Porn and Bigfoot

Search term data provides insight into how we react to the economy. They found that when there was a new high in the gas prices, there was a spike in visits to gas related sites. But the searches would taper off in a week, even if prices continued to rise. But over the last six months the searches have gradually been going up. He thinks this represents a lasting sensitivity to gasoline prices.

He's been quoted a lot about porn. He's found some interesting insights into adult Web sites. Over the last five years, every summer is the low point of visits to adult Web sites. However, there was no dip this summer. Looking at the demographic data for visits to these sites shows that households earning under $60,000 were visiting the most. His theory is that during times of economic downturn, people turn to diversionary sites like adult sites.

Search data can also show some of the popular Halloween costume ideas. Bigfoot is the most searched costume this year, and it could be because of the Bigfoot hoax a few months ago.

Stacy Keibler, Sarah Palin and "Hot Photos"

As he was putting together a presentation, he was watching a news story about Dancing with the Stars. It was the season where Drew Lachey, Jerry Rice and Stacy Keibler were the final contestants. Knowing that viewer votes count for half the score, he decided that the most-searched celebrity would be the winner. Stacy Keibler was getting significantly more searches. He predicted Stacy would win, but was wrong.

Trying to find out what actually happened, they looked at the breakdown of the searches. It turned out that many of the searches were for hot pictures of Stacy Keibler. It's likely that these searchers weren't actually voting. (Surprise!) These findings led to the Stacy Keibler Correction Coefficient. Because of market conditions, people were not searching for the intent the researchers expected.

"Searching for Palin's 'Hot Photos'" was a Time column that Bill recently wrote. The week after Sarah Palin was named to the Republican presidential ticket he started gathering search data. There was a huge spike that week in searches for "Sarah Palin" but the searches were very Stacy Keibler-esque. People were looking for "hot photos" of her.

His research shows that here are three segments that adopt online technology before the mainstream: Young Digerati, Money & Brains and the Bohemian Mix.

Watching Search Trends

Specific trends of the early adopters:

• Cam-based social nets
• Personalization
• Tattoos
• Deal sites (resurging)

The Google Chrome team asked Bill to do the same analysis on the browser. The findings showed that a group designated Executive Suites was the main group adopting the browser. The theory is that this is because Chrome is only available for Windows, and not Mac OS or Linux, while early adopters likely prefer Mac OS and innovators seem to prefer Linux.

There's a few ways to learn more if you're interested. Visit the Hitwise Intelligence Blog, www.ilovedata.com. Read his column on Time.com: The Science of Search. And read his book, Click: What Search Activity Tells Us About Society.

Posted by Virginia Nussey on 10/ 6/08 at 3:19 PM | Comments (1)
See more entries in Analytics, Liveblog, SMX East 2008

August 21, 2008

Post-Click Marketing: Converting Search Engine Traffic

Whose idea what it to have just a little snack break instead of lunch? I can't work like this. And by like this I mean covered in chocolate from my delicious ice cream sandwich.

This session the moderator is Anna Maria Virzi (ClickZ) and panelists are Carrie Hill (Blizzard Internet Marketing), Laura Wilson (New England Journal of Medicine), Scott Brinker (ion interactive) and Tom Leung (Google).

Our first speaker is going to be Carrie Hill. She thought she was going to have to bribe us with alcohol to get people to this session instead of SEO secrets. The real secret is knowing that Lisa's over there liveblogging it; it's just like being there!

Qualified traffic is the key to good post-click marketing. Buyers know what they want and that's what they'll search for. Use segments to deliver language and interface on those pages that will appeal to your shoppers. Use your trigger words. Buyer use words they relate to in their queries. If they use a word in their search, you should use those words on your page in order for them to see relevance. It should show up in the SERP and on the page.

Example: Free shipping-- Apple doesn't have free shipping prominently on their page so it's easy to over look. Zappos makes it obvious that they have free shipping.

Make sure that your visitors are landing on the right page. The home page is not right for every query. If they're using a word, give them a page that's relevant to it. Give your traffic the trigger word that they're looking for. If they do land on the home page, let them segment themselves.

Carry the message through the segmented path. IF they travel down the 'free' trigger word path, repeat that message.

Remember each piece of PCM can lead to more revenue from your site. Many pieces work dependent upon each other. Remember that halfway is only halfway but every little bit helps. Use Web site optimizer, do tests, let the users design their experience through self-selection.

Laura Wilson is going to present a case study on how this worked for the New England Journal of Medicine.

The Five Key Ingredients of Their Success:
1. Know what the audience is looking for
2. Engage and Convert visitors with relevant content and offers
3. Give the visitors a reason to come back to the site: Videos, beta site, free weekly audio summary and more.
4. Deepen relationships with the audiences: newsletters and subscriptions, information about updates to the site.
5. Optimize conversions through testing

Tactics:
1. Navigation links with calls to action: both home page navigation and global navigation
2. Offers on Sign In Pages -- offers that are relevant to visitor based on the content they're trying to reach.
3. Free trial upsell on the registration confirmation page -- after registering for the newsletter, they offer them a trial to the online version of the journal.
4. Offer in authentication string message -- offers based on level of access.
5. Targeted emails -- welcome e-mail series and a "new features" e-mails. Free trial member will also be getting an email series with a countdown on time left.
6. Promotions in Weekly NEJM E-mail table of contents.
7. Banner ads throughout the site
8. A/B testing, multiple tests

Scott Brinker is discussing segmenting. That's been the big thing this conference that I've noticed.

Two Takeaways:
1. To increase conversions have more specific landing pages.

A/B testing -- for your respondents, it's still just one page. You need to understand who your respondents ARE. Some might think that one thing is more important than others. What you need is more than one landing page to reach more than one audience.

2. Self-segmentation after the click

Some keywords won't give you intent. Have two step landing pages in those cases: "Dinner" -- do you mean "hamburger" or "pasta". You'll speak differently to small businesses than enterprise level pages. Tailor your second landing page to that self-selected audience.

Don't ask them to do too much work though or they'll bounce.

Figure out which ads attract which segment. Then see how well you're converting those segments.

5 reasons that 2 clicks are better than 1:

1. Easy Engagement - makes it easy for them to move forward
2. Self Identification - we respond to self-identification cues, more accurate than forms, sets expectation
3. More focused content - contextually relevant content sells better
4. Signaling - Investment reflects commitment. "If you target me, you much think I'd be a good fit..."
5. Market research - which ads attract which segments? Which segments convert best? How do prospects think of themselves?

Last to speak is Tom Leung from Google's Website Optimizer

In the old days, you just implemented stuff and hoped for the best. Or you listened to the "HiPPO" the highest paid person in the organization. If you were a little more advanced, you'd do a before and after test but that wasn't that enlightening.

Website Optimizers allows you to test different variations of a page to see which version is most effective at achieving results.

This puts power into the visitors and they'll tell you what they like best. Sites can be a living laboratory.

[He quickly goes through how to do testing with Website Optimizer.]

The only opinions that matter are the opinions of the people who go to your site.

Don't assume, make sure that your revisions aren't going to HURT your site. You have to test with a control. Your interesting idea might not work.

Basic questions:

• Does it look legit?
• Is it intelligible with partial attention?
• Is it simple to convert?

Advanced questions:

• Is it compelling?
• Does it handle top objection elegantly?
• Does it provide all the essential information?

If you're thinking about outsourcing:

• How many experiments have you run?
• Referrals? -- screenshots and contacts
• Can you justify ROI?
• What was the average lift?
• Can they work with your IT department?
• Are they willing to tie their payment to performance? (not required)
• Do they have marketing, proj management?

Ask yourself if it really makes sense to show ads on your landing page. Tell people what you're about.

He likes the Netflix landing page: It's clean, legit, informative and not too complex.

Q&A

How do you get buy in?

Laura: We present data and do projections on what the impact could be.
Carrie: We had to do a little bit of free work to show them how to make the lift. Sometimes one test isn't enough. But once you can show them the difference that a little work does, it's not that hard to convince them to do more.
Scott: It comes down to two things: Make the argument about conversion rate. Also web site optimization is a huge task. Landing page optimization is smaller and easier.
Tom: Agrees with Scott. Don't make it a huge plan, just do the simple A/B test and show them the results and the lift. People find it hard to disagree with more conversions for the same money.

How do you use Website Optimizer on your home page?

Tom: Put the goal tags in multiple places and all those are considered conversion OR they'll do a time on page test and consider that a conversion.

How long should a test run?

Tom: Never shorter than one or two weeks. Have about a 100 conversions per combination.

Is it possible to use optimizer against a segmentation page?

Tom: I've seen people run tests where A is the regular landing page and B is the segmentation page.
Scott: It's hard to answer that without sounding like a sales pitch but yeah, that's what our tools do. It's possible to do even with just a simple test. You can at least take a first step in that direction.

[Skipping an asked and answered question and a very specific question.]

[I don't know what his questions was but he said statistically relevance about ten times. I think it involved math. Tom's answer was all complicated and technical too. I'm sorry, I can't even begin to interpret. HOWEVER: Green = high confidence, Yellow: mid level confidence, Red = low confidence loser]

How do I test on low volume keywords?

Tom: Keep it simple. Just do an A/B test. Also, change your conversion metric. Make it time on page instead, so that you can take that as a leading indicator to conversion.

Scott: There's nothing wrong with A/B testing. It works.

Carrie: Don't get sucked into the idea that a conversion is 'they bought something'. It can be moving to the next step. You're testing a path sometimes.

Tom: I'd agree about the power of A/B testing. At the end of the day, people get the best results from very small tests. Small tests make you focus. Multivariate tests can make you lose your focus.

Posted by Susan Esparza on 08/21/08 at 2:31 PM | Comments (0)
See more entries in Analytics, Google, Liveblog, Pay Per Click / Online Ads, SEO Tools, SES San Jose 2008, Search Engine Optimization

August 20, 2008

Social Media Analysis and Tracking

No time for chatter--let's jump right in. Marshall Sponder (Monster.com) moderates, Breanna Wigle (Military Advantage), Edmund Wong (iCrossing), Todd Parsons (BuzzLogic.com) and Rob Key (Converseon) speak.

Marshall, who we met yesterday in the Analytics 2.0 panel, introduces himself and the panelists. He's got a short presentation to explain the session. It's possible to categorize referrers for social media attribution and find out how much is coming from where. Don't count hybrid sites like a news site with a social aspect. Stick to the pure sites for analysis. You can use ComScore to measure social traffic as well. Just use the conversational media category. It's just meant for research, not actual traffic but you can learn from it. He thinks that we should be able to let Google do it for you as well. Google already knows what sites are blogs and could segment data from a site from Social Media. It would be easy for them. Um, kay.

He repeats that he thinks that CMOs are going to be Web analysts in the future. It's so important to bake in analytics right from the very beginning. Right now Social Media has no set place in most organizations, ROI is difficult but not impossible to prove.

Web 2.0 is about empowering users. Measurement is in conversations/engagement, Traffic/ROI much more subtle but still measurable.

Rob Key is our first speaker. He utters my most hated phrase "I have a lot of slides and I'm going to try and go through kind of quickly." Noooooo!

Stats! Lots of them! 45 percent of adults have created content online. There are 1.2 million blog posts a day! Other numbers as well! [That's all I got.]

Social media is growing and infecting search as well.

How do you design a social media strategy? Listen to what people are saying, engage them, and then measure and optimize.

Listen: Conversation mining helps marketers promote and protect their brand through the measurement of analysis of online...something. Please slow down, Rob.

The conversation is iceberg like. The top twenty results in the search engines are what's above the waterline. You want to mine deeper than that. There's so many avenues out there that aren't ranking that you need to dig into.

Look at:
How are people feeling about our brand?
Who are the most influential voices?
How effectively are we contributing or not to the conversation
Who should we cooperate with?
And more questions as well.

You need to see who the influentials are in the space because then you can see who you need to partner with or counter. Your own contribution is usually about 10 to 12 percent.

What's the sentiment and the tone of the conversation by relationship? Positive, negative? It helps inform conversations and response.

Look at your cloud tags and look at it compared to the way that you talk about yourself. When you do this eh right way you can find out about yourself.

You can't just rely on automated solutions. You need human intervention when you're categorizing it.

How do you use it?

As an extension of customer service. With people complaining about products all over the Web, you can go in and use this as a marketing opportunity to salvage or turn around the situation. Avoid Dell hell type listings.

[About six charts fly by in 30 seconds. Oh help]

Where does this all go? We're still evolving. Trending is the bleeding edge now. It's not just data at the moment, it's looking at how the conversation is changing over time. We'll see some flattening of capabilities. Everyone's going to become better at this in the future. The other thing is that people will start to see the grand unified vision of data and you'll be able to get a bigger picture of your whole organizations.

How do you find the meaning the measurement? That's the next critical step.

Breanna Wigle and Todd Parsons are going to present a case study for Breanna's site: Military.com. They're going to be switching back and forth throughout the presentation.

Breanna goes through some background on Military.com--they reach out through blogs and connecting people. They're trying to enable communities to connect, not trying to create them. They found that social media traffic converts 6 percent better than non-social media traffic and have a higher time on site.

The goal of their campaign was to increase product awareness to new influencers and their audiences and to convert visitors into RSS and Newsletter subscribers. The challenge was the long tail and that it was hard to find the influencers because the fragmentation was very high.

They partnered with BuzzLogic for that. Todd says BL isolates and ranks influential content in social media across topics and serve ads within that influential content. They consider credibility, relevance and more in their algorithm.

Step 1: Uncover conversations.
Step 2: Rank the influencers.
Step 3: Identify influencer networks for ad placement and follow influence paths

In social media advertising, the creative is critical. It must be Compelling, Informative and have a Clear call to action.

Breanna steps back up. Results: 86 percent higher CTR compared to historical average for targeted banner campaigns. Conversions increased 5.3 percent in RSS subscriptions and Newsletter subscriptions. 90 percent were new visitors during the campaign, 60 percent higher than site average. There was a lift in time on site as well.

Todd--

Key observations and learnings:

• Active conversations about specific topics attract passionate audiences. Highly targeted display ads can perform well in this environment
• Social search is different than Web search and traditional site targeting -- it's about sourcing information via what "trusted" people are referring to. This can get you closer to "search" like intent.
• Influencers and their network relationships - the nature of linking connection matter when it comes to ad performance. Sites that connect to each other around specific topic are key targets
• Conversations offer a new window in analyzing user psychology and intent -- the nature of the conversation can impact as performance.

The more personal the experience, the more likely you are to subscribe to something.

Is there enough inventory out there? Todd thinks so and that it's mostly untapped. The question is finding quality inventory.

How does it compare price wise? It varies. In general it's less expensive to advertise on blogs.

Edmund Wong is up next. He's looking at developing a social media engagement measurement framework, looking at the customer experience and how to improve it.

Another case study. Tech forum engagement. Their goal was to be useful and helpful, not to sell a product and instead overtime improve perception of the brand. Natural search ended up driving significant long term brand impressions on accumulated postings. Measurement is key but standard ROI measure doesn't work in social media. Measure things that make sense for the campaign--postings, tonality, number of links, amount of traffic from the links, number of conversations engaged.

Quote from Clive Thompson: "Google isn't a search engine, it's a reputation management system."

Since many forums are optimized for search, engagement is highly visible. Forever. Traffic increased over time from the posting a year prior. SEO continues to drive new page views.

They took a look at what they COULD address and what they COULDN'T. They didn't fight battles that they couldn't win whether it was 'what should I buy' or customer sentiment issues.

They used it to identify issues on the site as well and were able to correct customer confusion.

Key Takeaways:

• There's no one killer metric for social media
• Track anything possible to glean insights
• Not just about numbers
• It's all relative (focus on benchmarking and trends)
• Measuring social media does not equal ROI for social media
• View social media as the world's largest focus group.

Q&A

How do you get people to join your group on Facebook?

Edmund: You have to figure out what your end customers actually need and what you can offer them. Get out there and show engagement. What's the useful and interesting aspects.

Rob: Ask not what the community can do for you, ask what you can do for the community.

What do you do with imposters?

Todd: Did you out the imposter? Yes? Then the social media democracy will take care of it.

Rob: Look at Jeremiah Oywang's post on companies on Twitter. Transparency is key.

Posted by Susan Esparza on 08/20/08 at 4:02 PM | Comments (3)
See more entries in Analytics, Branding, Liveblog, SES San Jose 2008, Social Media

Searcher Behavior Research Updates

And we're back from the fastest lunch I've ever eaten. Where does the time go? Moderating this session is Bill Muller (iProspect) and our panelists are John Marshall (Market Motive), Pavan Lee (Microsoft), Dr. Larry Cornett (Yahoo! Search) and Bill Barnes (Enquiro Search). I have to confess I just love the research sessions. Hard data just makes my little heart sing. Come on. You can't tell me you aren't excited about this one too.

I know you are because Bill is telling us about how every year this session totally fills the room it's in. Why? Because if you know more about the way searchers behave, you're going to be a better marketer.

John Marshall starts us off.

The interesting thing about search behavior is that it's not that difficult to get good data. The question is on Monday morning, do you understand search behavior? Most people turn to the keywords report in your analytics tool. That's a reasonable place to start but it's an extremely narrow view of the activity on the Web. You're only look at the keywords that brought people to your site. You only see the search results that brought people to your site. You're running into sample bias.

How can we really see the intent of people, not just the people who made it to your Web site. You don't want the whole forest view. You're probably not going to get the whole forest view anyway unless you have a Hitwise account or something. What you can look at is the single tree of your site.

The trick is to use the site search on your Web site. If you don't have site search, implement it. Even if it doesn't work, it's a great source of user intent. Search engine keywords only give you the people who came to your site. Site search gives you the intent of your users, conversion rate information. A lot of people ignore this data because it's free. Free data is often ignored. If you pay for something, you value it more.

Things that can go wrong:

• Mixed Case -- Google Analytics doesn't automatically change case for you so your data gets scattered across case. You need to convert it.
• Multiple results pages -- Some site search pages for 'no results found' don't get tracked by analytics. Make sure all the pages contain your analytics tracking.
• Usual JavaScript breakage
• Injected terms -- Most Web sites that have site search, they use it as a cheap landing page creation system. You have to filter that out of your data if you're doing that because it's not real data. No one is typing it in.

By using site search you're answering the question: what's the true intent of the users when they're on the site.

Site search data cannot replace competitive analysis but it's the cheapest way to get good data fast.

Pavan Lee is up next. She's from Shanghai.

Background from the New York SES: They've discovered that search listings have a branded value. Paid search listings have a stronger branding impact than organic search. There is a positive branding effect for both. They're trying to measure the brand lift.

They studied five brands in five spaces.

Methodology: Eyetracking and post-search survey.

Key findings: Search display and content ads are effective for branding stand alone but more effective together.

They asked "did you remember seeing an ad" 21% lift content 30% display. 38% both.

I can't see her slides at all.

In all cases with all questions, including lift in purchase intent, there was a brand lift and it was stronger for paired ads.

On the eye tracking side, search is still the most effective tool in attracting attention. There's a roll over impact on a multi-channel exposure. If you see search and display or search and content or search and display and content, it's more effective than just seeing any one of those.

Key takeaway: The power of three. There's a synergistic branding impact across content, display and search ads.

None of this data is public information.

Larry Cornett steps up to the podium.

His talk will build on John's presentation in a lot of ways.

Users do a lot before and after they're on the search page. He's going to talk about that, about the research they're doing, how users experience search, a little about crafting search and how they get from 'to do' to 'done'.

The reality is that the search page is just a tiny slice of online activity. Before the search, the user somehow comes to need to do a search. After the search, they want to go somewhere. They're going somewhere because they want to fulfill a task. The task is not getting to best buy. It's getting an iPhone. You need to know what happens after and how it all links back. How do you support them through the whole lifecycle of what they're trying to accomplish.

There is no single methodology that gives you the whole picture. Some ways that Yahoo does testing are:

• Search editorial
• Bucket testing
• Metrics & Analysis
• Search Science
• Focus Groups and surveys
• Eye-tracking research
• Ethnographic studies

How users experience search

• Starting context (what have they seen and experience before they query)
• Quick Scanning (one to three seconds)
• Information Scent
• Matching intent
• Quick Decisions
• Looking for answers (not a homework exercise. "Don't make me work")
• Feeling safe

They try to help crafting searches with 'search assist' (suggested searches). For most people search is hard. They're not experts.

Focus on the ultimate goal. They're looking to do something, they're wanting an answer. Yahoo SearchMonkey is an attempt at giving them that answer. It gives the user more information about what's behind the link and what's important to know.

What does this mean for marketers?

• Before the SERP
  • Starting context
  • The "Real task"
• On the SERP
  • Intent and information scent
  • Searchmonkey
• After the SERP
  • Fulfilling expectations
  • Being their "answer" and living up to the promise of the search result.

We thought that the reason people were having trouble with search was that it was an artificial session. But field studies showed us that the users were really having trouble formulating queries so we really tried to implement something that would help them.

Bill Barnes is the last to speak.

Their research is grounded in their search marketing and grew out of that.

Why is the first listing seen so important
Why do we scan in groups of 3 or 4
Why branding is important

[Standard heat map image, you've seen it a million times.]

They did experiments with the top SPONSORED listing and played around with really great ad copy versus just 'okay' ad copy.

When they did a survey they didn't ask about the listing, they asked about the search engine and if they'd use it again. The only difference was the ad copy but there was a huge lift in trust in the engine with the great copy.

Working memory: It's what comes to mind with recall. For some reason, we're hardwired to think in threes or fours.

[Oh no, my battery is dying]

There's a 16 percent increase in brand association when brand is the Top Sponsored and Top Organic Results. On an unbranded query. The really interesting thing is that the recognition of OTHER brands drops away at the same time.

There's an 8 percent lift in brand purchase. If you're not there, you lose 16 percent brand lift.

Even for branded queries, you get a brand lift if you appear. Should you buy your branded terms? Yes.

Eyetracking finding: Brand fixation only occurs in the TITLE and the URL not in the description.

If you're a familiar brand to the searcher, they will often skip the sponsored listings at the top. If you're buying the top sponsored, write your copy for a NEW user.

If you have brand A and brand B in sponsored with Brand A in top organic, brand A gets a HUGE lift.

Key Findings:

• INTENT is the most important thing
• Organic and sponsored combined give the biggest brand lift.
• Be aware of who else is on the serp
• Write your ad copy to new clients.
• Don't assume your brand will be in the consideration set. If you're not on the page, you're forgotten.

Q&A

The first question is does offline affect offline. The answer is yes, though the panelists don't say that. Go read the Re Search Online, Purchase Offline session from yesterday.

Why do search views get longer?

Pavan thinks it's because searchers are looking for something in particular whereas display and content ads are push forms of advertising.

Do they really only spend 1-3 seconds and how often do they click?

Larry: It's on average. In some cases, for navigational queries, that's less than a second. It might be longer at home but yeah, it's amazingly fast.

Bill: Females look longer and shop around, males just go straight to results. There's a free paper available.

Pavan: Search intentions lead to searcher behavior. Fact based search stays organic. Commercial searches tend to be more broad. It also varies by culture. Chinese spend twice as long as Americans.

John mentions that his contention is that the site search queries are the same queries that are being typed into the search engines but they're just not getting to your site.

Is the suggested search condensing the search queries?

Larry: Yes. People are moving to longer queries and that search assist does jump them to the queries that will get them to the answer faster.

Are there differences in lift by categories?

Pavan: Yes there is a difference in lift across different verticals but in all cases it does result in lift.

[Long set up about pretending to be a confused searcher and poor SERPS] What can be done to help confused searchers?

Larry: Search assist is just one way. It works mostly for shorter queries?

Does the golden triangle change with non-roman character sets?

Pavan: In Chinese, the scanning pattern is very different. It's a rectangle. You have to look at everything to put together meaning.

Would you suggest not trying to dominate the organic?

Bill: No, never. Always optimize.

What plays into search assist? How does it affect PPC?

Larry: Nothing is paid in those.

John: The hidden message there is: No you can't spam the suggestions.

John says that the other thing site search is good for is manifesting usability problems and for doing competitive intelligence.

If you rank 1 on a non branded term, should you also be number one in paid search as well?

Bill: That's exactly what our research showed. That said, always test and retest and see if the ROI is worth it. Clicks went 50/50 on paid and organic, so make sure that you're testing and monitoring.

Posted by Susan Esparza on 08/20/08 at 2:10 PM | Comments (0)
See more entries in Analytics, Live Search, Liveblog, Pay Per Click / Online Ads, SEO Tips & Tricks, SES San Jose 2008, Search Engine Optimization, Yahoo

Keynote Roundtable: Why Does Search Get The Credit For Everything

I hope everyone had fun at the Google Dance last night (just try and take my light-up Google ring away from me. I'll kill you.), but now it's time to talk search again. It's also time for a keynote roundtable. This is where smart people come to unleash search insight and talk over one another and Lisa's finger's fall off as she tries to keep up and decipher it all. I wonder who I pissed off to keep getting assigned the morning sessions while Susan and Virginia sleep? Oh yeah. I pissed off Susan. I hate her.

Kevin Ryan (SES, SEW) and Bill Hunt (Global Strategies International) are moderating Randy Peterson (Procter and Gamble), Mikel Chertudi (Omniture) and Sharon Gallacher (Neo@Ogilvy). Let's do it.

Bill says when we look at the media spending mix we see lot of things going on. 96 percent of the budget goes to things other than search. For many large companies, this is the media mix. For many dotcoms, the media mix is exactly the opposite.

Bill says he gets to sit in a lot of meetings. He hears all sorts of silo'd marketing tactics. He hears that search is driving $250 million in revenue. Print is generating $100 million. Direct is generating $20 million. Then, at the end of the year, the CFO jumps up and says we only sold $100 million so someone is double counting their numbers. He wants to know who it is. It's a matter of attribution.

Randy: When it comes to attribution, you have to think about it not just in terms of sales value or direct response, but also in terms of branding. You have to think of the brand value that's created by search. When you build the brand there's value associated with that.

Sharon: The branding aspect is very, very important. If you can make a 5 percent increase in branding, then that's a big win.

Mikkel: In addition to the sales and orders and lifetime value, it's important to realize that when you take credit for sales, you also have to take credit for the cost.

Bill talks about the "last click" concept.

Mikkel: There's a display team responsible for all the banners. A user clicks on that banner ad and then they click on the email that comes in their inbox. Then they go and do a search in Google on the keyword [auto insurance]. They finally click through and fill out a lead generation form for a free quote. Many companies are just giving credit for that last action. What should be happening is that they should be measuring and accrediting all the actions that came before. What started it, what came between, and where did it end? You want to create a baseline on all of those. You can't ignore the other channels.

Kevin: Are people understanding that search is reactive in a lot of respects and we have to have to figure out how to tie it back?

Mikkel: We see that there's a lot of brand being driven by TV and display advertising. We see search gets credit for everything. There's also another issue - cookies and cookie settings. Sometimes cookies expire in 7 days or 90 days or never. If you click on a banner with a cookie expiration date set for 30 days and then you click on an email link 31 days later, the display ad never gets credit.

Randy adds that comScore has found that 30 percent of people delete their cookies once a month.

Bill: What are some of the things that can go wrong if you don't attribute properly?

Mikkel: He worked in display advertising about 8 years ago for the Web's 2nd largest publisher site. They had two separate installations of an ad server, which caused them to double count everything. They were a direct response ROI-based business and found that they were trying to tie their cost-per-customer down a very specific number. When he would buy ads on a CPM basis they found that they had overspent to the tune of about 20 percent aka millions of dollars! That's because they were over-assigning credit to the search channel and to the display channel.

Bill: This whole idea of attribution leads to this unhealthy tension between search and traditional advertising folk. People look at search as the Marsha Brady of marketing. It can lead to turf warfare and unhealthy tension. Are you guys seeing this?

Randy: It's still early on for us. Search is still somewhat in its infancy. They attributed too much value to search and got unhealthy tension. They went back and assigned conversion goals to activities happening on the Web site so they could associate values. They're building a model to do that and that has helped a lot with the brands that are spending big money on search.

Bill: Talk about this idea of misguided investment.

Sharon: It can take you a long time to ever find out that you've invested poorly, and that's scary. If you're off and you keep being off, you just keep building up.

Kevin: Is there a lot of resentment internally?

Randy: There's still some of that, yeah. The strength of personalities in large companies makes things happen. The way around that is through data. Data beats personality.

Sharon: We're seeing that in the rise of display and search. The TV budget was 3x the search budget for a long time. Having the data just obviates that argument.

Mikkel: If you have the data you can outmaneuver the strongest will in the office. Hopefully all of us in the room have a bit of a competitive advantage because we could manipulate that data. We're not doing that, of course.

Bill: We want to talk about some of these metrics. We may be measuring the wrong things. Attributing results to the wrong things. What are you measuring? What are your pre-metrics (impressions, clicks, CPC) and post-metrics (orders, sales, lifetime value)?

Randy: You do a bunch of marketing activity and bring some value. Some of the metrics we think about are anything we can collect from analytics. That's the broad category. It includes clicks, data we can pull from the engines, ad serving data, etc. We also look at all offline media. You have to put all that together. In the end, you also have to look at sales, searches on an ecommerce site, a major brand value, etc.

Mikkel: He likes Randy's buckets.

Bill: We had a client who found that about 65 percent of their traffic was coming from organic, and 40 percent of that number was engaging with the live person execution they had on the site. They ran a report and decided that they should decrease the amount of search. They couldn't tie back any sales from the traffic from search. Many people overwrite the conversion source. If someone comes in from search and then talks to a sales person, the sale's person's code gets put in for the conversion, not search. Can you talk about that?

Mikkel: This is a really important issue. You don't want to overwrite objectives. Your objective is to bring in visitors to your Web site. Your site objective is to drive conversions. The remarket objective is to get repeat orders and cross sell. It's critical not to overwrite because you're overwriting your costs for those as well. You can't do full blown ROI analysis that way.

Randy: There's a quote that says all models are wrong, some are useful. In your case, the model was wrong. What you did was made the model more correct and then it became useful for you. That's a good way to think about this.

Mikkel: The takeaway is don't overwrite. Keep them all in the database.

Bill: Most attribute success to the last action and all decisions are based on these metrics. Is this wrong?

[A few people in the audience say it's wrong. A few more say it's something we have to live with. And everyone else is just hung over from the Google party.]

Bill: Is it possible to set up a fair and balance program that gives credit where credit is due? He says it is.

Mikkel: Sort campaigns by an array and look at which was the first campaign that brought this account into your database. What was the last one? What is the secret sauce that worked to get that campaign into your database? Once they start to repeat purchase, you want to track that independently as well. What was the last campaign before the repeat purchase? Store and measure all the data.

Bill: If we solve the technology problem, can we really change the hearts and minds of fellow marketers?

Sharon says if we solve the tech problem, we'll solve the attribution problem somewhat. We'll never get it solved 100 percent but we can find a better balance. There are different measurements we can use. They're not all the same thing. Maybe we just have to daisy chain those pieces of research and make assumptions.

Randy: He agrees with Sharon. Data is so important. It's friendly in this situation. It brings the discussion down to something that's less emotional. That's how you win over the hearts and minds of marketers.

Mikkel: Marketers find peace in the numbers. He can justify his job in the numbers. It's important to be data-driven marketers on the metrics that matter.

Tips for Success

• Cover the basics: Do you know where this data is?
• Define your goals effectively: What are you trying to manage.
• Define the value of the action
• Know your bounce rate: We hear about the 5 percent click rate. What happened to the other 95 percent?
• Look for tools or indicators that you have in other media to help indicate impact and success.
• Create an attribution model, using data available.
• Get comfortable with not having a perfect answer.
• Standardize tracking and marketing applications on one system.
• Measure all views: 1st, last and shared.
• Don't overwrite metrics and objectives: Track them separately.

Posted by Lisa Barone on 08/20/08 at 10:24 AM | Comments (0)
See more entries in Analytics, Liveblog

August 19, 2008

Identify, Analyze, Act: SEM by the Numbers

Wait, SEM by the numbers? Oh noes. How did I end up in this room? Lisa doesn't do numbers. She's pretty. And talking about herself.

Moving on.

[And Li Evans just snuck up behind me and gave me a full-fledged heart attack. Thanks, Li! :)]

Chris Boggs (Brulant) is moderating speakers Craig Danuloff (Commerce360, Inc.), Brian Cosgrove (AvenueA/Razorfish, SearchMarketingGurus Author), Heather Doughery (Hitwise), Michael Stebbins (MarketMotive) and Brett Crosby (Google).

Up first is Craig Danuloff.

Ooo, he has Superman on his first slide. Susan so wishes she was here.

There are three things you're up against when dealing with Web analytics: Invisibility, deception, and unlimited power & resources.

Invisibility: You miss the full spectrum of what's going on because the start and the end of what you're measuring are both missing. Every search is a question; every ad is an attempt to answer that question. Keywords are simply a connector. ROAS is a "feel good" metrics that you shouldn't take seriously.

Deception: Can you really trust what you see? You have tens of dozens of campaign groups and all that ever comes back are the averages. You see a rollup of your average position of your profit and ROI. It looks good but if you haven't seen the raw data, you can't make informed decisions. You also have to worry about accuracy. You don't always know if you have a statistically significant sample size.

Unlimited Power & Resource: You run huge campaigns and time is an issue. You have to watch where you put your energy and your effort. A change is not a test. Someone needs to be there to record it, to track it and to alert you of the results. Not all the tools provide that, but you have to be more methodical about watching it. You have to manage inter-dependency.

How Good Can Triumph Over Evil

• De-Cloak 'em: Request a way to get information on queries. Demand real margin-net Profit-ROI numbers in SEM Analytics
• Keep 'em Honest: Constantly segregate keywords by performance. Understand revenue and expense attribution and allocation. Don't get lied to by statistics.
• Use the force: Apply math and statistics to numeric problems. Make changes with testing and inter-relations in mind.

Good stuff!

Next up is Brian Cosgrove.

Implementation: If you don't configure your analytics package correctly all of your referrals will come in as organic search. It's just going to bucket it that way. You need to make sure you configure it so that you're separating everything out, especially if you're using something like Yahoo Paid Inclusion.

Filtering: Lots of people don't take into account that they have a lot of internal traffic happening. People come in for different reasons. You want to put filters in there. Go through your reports and look for things you don't understand.

Data Driven: With all these different platforms you have lots of people sending in ad hoc requests asking you to get numbers or reports on something they need. They're not using it to drive business decisions. To get to a data driven organization, changes need to happen.

Roles: Web analysts are very good at looking at data. They have a general understanding of what they're seeing. They come up with insights that they have to feed along to Web developers. But what they're missing is an operations person. There needs to be a Project Manager to keep everyone on task and make sure things are getting organized and done.

Process: Report -> Analyze -> Optimize -> Measure. This process gives you full utilization. Analysts are always putting stuff together and people are always ready to take them.

Create landing page reports. This will help you to find the pages that happen to rank well but where the word isn't exactly the best place for someone to make their way in the funnel. It's important to look at the different words and see which landing pages are really working for you.

In conclusion:

• Implement your platform correctly
• ID actions you can take ahead of time
• Coordinate with other resources
• Separate the analysis cycles
• Staff people to manage the projects

Heather Dougherty is next.

Identify and Maximize SEM Opportunities:

Identify trends and seasonality: Be proactive by planning ahead for trends. Keep in touch with that's going on in marketing. Understand the reliance of paid vs. organic search within specific industries and do the same for any competitive Web sites. Measure the impact of brand awareness upon competitor's organic traffic.

Compare where people are searching to where they are clicking.

Improve keyword list efficiency by breaking out the best terms for optimization vs PPC campaigns.

Identify who is doing well in sponsored listings and learn from their copy. What are they saying that works? What's in their copy? How are they attracting people?

Learn from the best optimizer. See who's getting the most traffic and figure out why they're getting it. If you can't beat them, maybe you can partner with them (ala Google?). Find good partners for product placement, affiliates, etc.

Determine user intent - purchase or new? Can help you optimize for that strategy?

Monitor your brand health and use search for reputation management in a crisis.

Integrate search findings across the organization. Do research to determine the best search engine optimization and PPC opportunities. Take advantage of these findings to help drive all marketing initiatives. Identify advertising opportunities on Web sites that are well optimized for organic results.

Discover affiliate partners that are bidding on competitive terms to maximize your budget.

Michael Stebbins is next.

What's In Your Data?

Basic: Bounce Rate, Average Time, Page Views
More Vital: Conversion Rate, Cost of Visitor and Revenue per Visitor.

The Grim Reaper

1. Tactical Question: Which 10 percent of my ads are not performing?
2. Possible Answers: Find ads with high cost, bad ROI, low engagement, low conversion, etc.
3. Planned Actions: Cut them. Create replacement ads. Increased investment in the areas that are performing.

He runs us through a sample of The Grim Reaper technique. He also mentions that CustomerIntent.com has a calculator to help you figure out your margins, so you should check that out.

Take a look at your keyword stats and see how they're performing. Based on that, mark the candidates for deletion. You want to get rid of the campaigns that are not performing. Don't try and fix them.

In Google Analytics you can sort by ROI. The ones on the top (aka the ones with the highest ROI) you can ignore. The ones on the bottom are your candidates for deletion. Take a look at the number of visits, the ROI percentage, etc, and decide which ones should be deleted

Free Tools Love

• Microsoft's Check Commercial Intent Tool: You put in two different terms and it tells you, in percentage, the commercial intent for that tool.
• Google AdWords Keyword Tool: Tells you the search volume for your keywords with estimated CPC.
• Microsoft's Forecast and Demographics Tool: Tells you who is buying, gives you age and gender information. Helps you to tune your message to your demographic.
• Google Ad Planner: You list the sites you think are most likely to attract that audience and the tool returns sites that have the traffic you're looking for. (I have no idea what that means. I'm hoping you do.)

When you're doing your testing, rig the election by protecting the incumbent. You want to create three copies of the incumbent ad that's performing. You don't want to cut a performing ad in half. Set your ads to rotate evenly. This will give 75 percent to your incumbent ad and 25 percent to the challenger.

Brett Cosby will finish things up.

Brett says that today is the 4 year anniversary that Google approached him about buying Urchin. Aw.

When Google bought Urchin, Brett realized that their audience had changed drastically. They launched a new interface for Google Analytics where the goal was to put data into context. You get the big overview first and then it gets broken down as you go down the page. They kept the deep pool for the experts and added a shallow end for everyone.
Get the right data to the right people.

1. Set up goals and funnels (ecommerce and non-ecommerce).
2. Customize dashboards.
3. Customize your email reports for different roles.

Goals and Funnels: You have funnel reports - traffic goes in and you see how it leaves. Set up your goals via your admin interface. Enter goals, funnels and value. Once you do that, Google can tell you the goal value per visit and other data. Set up actual ecommerce for even better data.

Customized dashboards: You can put any report you want on your dashboard with a click of a button. You can also email reports - either instantly or scheduling them. Get customized data going out to the right people at the right time. Do this for each major role and everyone is happy.

Chris Bogg finishes things off by asking the panel what the audience can do now that will save them money the fastest. Here are their answers:

• Click the revenue data
• Figure out which terms you're going to optimize better
• Segregate your brand terms and look at the ones that are not brand terms.
• Look at your site and see what kind of duplicate content issue you have.

Posted by Lisa Barone on 08/19/08 at 5:49 PM | Comments (1)
See more entries in Analytics, Liveblog, SES San Jose 2008

Measuring Success in a 2.0 World

First panel of the day and I'm elbow to elbow with people eager to see what's coming. Andy Beal is two seats to my right snapping pictures like a paparazzi.

Richard Zwicky (Enquisite) is moderating this session with panelists Jim Sterne (Web Analytics Association), Matt Bailey (SiteLogic), Avinash Kaushik (Google) and Marshall Sponder (Monster.com). This is going to be a great session so let's jump right into this, shall we?

Avinash Kaushik is up first. Thanks to twitter he's now going by @avinashkaushik. Hee. He shares some pictures of his kids. They're so freaking cute.

Why is 2.0 such a challenge? [Slow down, Avinash!]

Content Creation, Content Distribution Content Consumption. The current flow is broken. Right now the BBC creates the news, puts it on their site and then people read it. You can track that. But the problem is now that you've got people other than the BBC who are posting on the BBC and distributing their content and that guys is also posting on his site and distributing it through RSS and it goes everywhere and now you can't figure out where the content originated and you can't track anything.

It's different than it used to be but he thinks it's better now.

How to track this new world?

Multiplicity. You have to use many different tools to track each different distribution channel. You can't build a house with just one tool. You need many different tools.

Unique Measures for a New World: it's more than just time viewed and traffic and unique visitors, but you need more than that. What he thinks is great is RSS. It's permission marketing at the best. You can push content to people who want to see it. He tracks feedburner every single day, not really the actual number but the growth. That's a critical metric in the 2.0 world. Think of unique measures that are more relevant.

Unique Data Collection: There are new things out there that break the usual tracking. Gmail is one page view. That stymies a lot of analytics. Hovermaps where you don't get a click, still one page view. How do you track all that. You need to look for a different way to track and measure. It's events not page views.

Jim Sterne follows Avinash. He speaks slower. Yay!

Web 2.0 is now Web 2.0.1. Things are changing a little bit. You site is a window into visitor behavior. There's data everywhere. You can track almost anything anywhere. Videos can be tracked not only for views but for where the rewinds happen, when the viewer abandons the video, etc. Web metrics is really growing up. It helps you get into the hearts and minds of your

Search metrics is on the same path. Ranking to Traffic to Analysis to Dynamic Bidding to Predictive Buying.

The little hitch is the economy. People are cutting spending. They're focused on maximizing profit.

We need to understand not what brings leads, traffic, rankings. We need to understand what brings PROFITS. You need to connect everything or your budget will be cut.

Know what your goals are.

Short and sweet. Awesome.

Matt Bailey jumps up next. Hi Matt! He talks really fast so I'm sorry that you're going to miss this entertaining presentation.

Captain Kirk is analytics GENIUS. Green alien women were the key to be safe.

Apparently this is the first sexist analytics session we're going to have. Oh Matt.

Analytics 1.0 is pages of charts and graphs and it runs you down.

Anlytics 2.0 is about questions! Something logical, something creative, something software can't do. Asking how, why, what if? You need to look

Context, comparison, contrast. Ask WHY?

Take an assessment of the situation. 450 people, five year mission, 54 total deaths (that's our conversion rate.)

But that doesn't tell you anything about the mission itself. You need to segment.

Death rate by shirt color: Yellow shirts die 11% of the time, Blue 9% Red 79%

But that still doesn't give us context. We need to look deeper. Keep segmenting. Segment the segment. Onboard deaths versus planetside death rates. Now we're getting a little closer to where the conversions (ahem) are happening.

Bring in another segment. In the episodes where Kirk meets a woman, the death rate drops to 16 percent.

People are not cows, they don't move through the site in a herd. They come to your Web site for different reasons through different keywords. Which means you need a different conversion rate for every segment. You have to know what they're looking for. Context is everything.

Telling a story is the only way to compare and contrast.

1. Ask questions
2. Take action.

"Question-asking is the most significant tool human beings hands."

Marshall Sponder steps up to the podium. He gives us a little bit of his history. He's from an artistic background and he didn't bring any slides.

They're studying social media. Social media standards are coming later in the year from WAA.

Social networks get their traffic from social media not from search.

Web 2.0 is about empowering people to contribute. The way that you can empower people on Facebook, etc, is giving them tools to communicate.

Social media traffic is more directed than search traffic. Unlike the traditional idea that search traffic is the most directed but it's really social media. The bounce rate is high, but the long tail is where the traffic is.

Look at the data like an artist. Focus on the big pictures and the patterns. There's a ton of data coming at us all the time. Try to answer the question that sparked the report in the first place. Try to figure out what companies need and answer their questions.

You need to have something for people to do when they hit a site. It's not just getting them to the site, it doesn't just stop there. You have to keep going and lead them.

Search is part of marketing but social media isn't really part of anything and it's not really clear where it belongs. It's hard to qualify the ROI from that. At the end of the day the CMO is going to be the Web analyst. Measurements are going to be baked in at the very beginning. You won't roll out things that you can't measure.

Q&A

Social media traffic, how do you segment it?

Marshall: We look at the logs and see where people are coming from. I want Google analytics to take into account ComScore data.
Matt: Again it comes down to segment the segment. You can't treat all social media traffic the same. Go back and see 'why are people coming here'. See what was said on Twitter or on Facebook. Set up the context. Here's what they saw as they clicked that link, now you know what people were expecting to see when they clicked and how well it matched up with what they got.

Page views, are they dead?

Avinash: They'll continue to be relevant for a while, most of the Web is still flat, static pages. We need to measure more that that a page loaded in a browser. We want to measure interaction. Page views will be with us for a while but they're not what we're looking for.
Jim: What he said. What I want to know is why someone came to my Web site. Viewing a page was not their goal. I want to know what it is.

How do you engage around What If?

Matt: You're off charts and graphs as soon as you ask what if? You're into the realm of creativity. Now you're into event trapping. It's not just about how long. When Vista came out, the average page visit was high but do you think they were having a good experience? You need to think about what it means and come up with creative questions about 'what if' then think of ways to test it.
Avinash: Before, the problem is that things moved so slowly that you had to torture the data to get it to confess. Why do you have to do that? Why not just do it? "Will this convert? Let's look at the data, blah blah" Why? Just do it. Send out a test email. For a lot of questions, the Web reduces risk, you can just try it. You don't need to spend a lot of time data crunching.
Jim: Pick your bumper sticker: Don't Make Me Think. Always Be Testing. Those are the core tenets.
Marshall: You can do some what if's now but you'll need the high end tools.

How do you integrate all the different data? Are there tools to do that?

Avinash: All the vendors are moving in that direction but if they tell you that they can do it right now, they're probably lying. Now your best friend is probably Excel. He also likes Crystal Excelsis.
Marshall: The more important part is the analyst, not the tools. Spend the money on the analyst, that's where the value is. Data is just data.
Matt: That comes up frequently, why are all the tracking data different? The question isn't why are you more accurate, it's why are you less inaccurate than everyone else. Don't take the best numbers out of which ever tools, that's a no no. You're looking at the trends, not the hard numbers.

How do you measure lift from dynamic bidding?

Jim: Keep working at it and when you perfect it let me know and I'll write another book. [hee]
Avinash: It's difficult to model human behavior that well. Dynamic bidding works more efficiently for long tail term. Head terms are much harder to model.
Jim: The long tail is where your most valuable traffic is anyway. When they're specific they're buying.

How do you find the why?

Avinash: the only way to find the why behind the what is qualitative feedback. Even something as simple as a four question survey is helpful. "Why are you here?" "where you able to complete the task" "Where did you have trouble?" [that's only three. I don't know where the other question went.]
Marshall: Didn't Google Analytics integrate something with question data?
Jim: You need to find why they went there. Then how they feel. Then what did they do--what was the outcome. Finally, you need competitive intelligence.
Matt: That's where the convergence of usability and analytics come in to me. I get more out of observing them. We can see where the hesitation is and back it up with the data.

Microsoft intregrated spreadsheets, will Google?

Avinash: The Microsoft AdCenter plug in is great. I'm very fond of it. The reason you keep using excel is that the tools available aren't that great at doing the segmentation you need. Google docs are being integrated into G Analytics.
Matt: Excel is my most favorite tool. It's fantastic stuff. You can't present analytics in power point. Excel allows you to display the relationships.
Jim: Every major tool exports to Excel because that's where the power is.
Marshall: Everyone uses Excel.

So we have to ask the right questions and get the answer but now what? How do I convince someone who is a manager looking at the Omniture data who only sees search views are going down and doesn't care that the page views are going up?

Marshall: (For Monster:) Search traffic comes in peaks and valleys. It's driven by spend and inventory. But the traffic continues. You need to understand the relationship. Fine tune it. You'll have to move up the food chain.
Avinash: Blog post: "Seven Tips for creating a data driven boss". There's a great report in Google Analytics: Visitor Loyalty and Recency. Measures not only what happened in the visit but across the visits and the distance between the visits. Segment first, then go in and see where these people are and where they're coming from and how often they're coming from. You can also survey the customers.
Marshall: All web analytics is really proving hypotheses. You might need to do more. Do comparison. Figure out what the right question is.
Matt: I think you'll see that the two points of data you have isn't enough. You need more. You need to go in and do more. In order to clarify a chart, add more information.
Jim: It also depends on goal. If the goal is just 'bring more traffic to the Web site through search' that's an artificial goal. You need to figure out the business goal behind it.
Marshall: Search has to go all the way up the food chain. If you're doing optimization and measuring afterwards, you don't have enough control. You need to base your deliverable on things that you can control.

Posted by Susan Esparza on 08/19/08 at 12:14 PM | Comments (0)
See more entries in Analytics, Liveblog, SES San Jose 2008, Search Engine Optimization

June 19, 2008

How Much Testing Is Too Much?

I was reading Search Engine Land yesterday (as I'm known to do) and came across Jon Miller's article how about much landing page testing is too much. Jon offers up a highly fancy-looking equation that you can use to determine when you're done testing your landing page. It made me giggle. A landing page test calculator? Really? There are three very easy ways to know if you've done enough testing and none of them include a calculator:

1. You've reached the burn rate where the test findings will breed only minimal results.
2. When there are other, more critical elements or pages to test.
3. When the person in charge of your testing process threatens to quit if you make them run one more darn A/B experiment.

See how easy that was? And no formula involved!

Truthfully, whether you're doing A/B Testing where you test one element or a time or you're using a Multivariate testing tool like Google Website Optimizer, you can test forever. It doesn't matter if you're talking landing pages, new site elements or better copy on interior pages, the iterations can go on forever. There will always be a better way to word something, a better place to put that image, or a new angle that your customers may die for. But at some point, the benefit of conducting the test looks like this:

(I have mad MS Paint skills, eh?)

Most search marketers will know intuitively when they've reached the burn rate for a particular test. It's when you implement the change and find it didn't improve the user experience on your Web site or result in a lower cost of conversion. When you get there, move on and start testing another page or site element. There's no use having one perfect page if the rest of your site is falling by the wayside.

When you're doing your testing, conduct it in one week increments so you take into account any day of week or time of day effects. Your test may run several weeks long or a even solid month depending on what you're testing. You shouldn't end your test until you have complete confidence in the answer it's giving you. Sometimes that means running something more than once.

Make you sure you have enough data when you're conducting your tests and don't ignore your baseline measurements. Your customers may have preferred the "old" way to the new one. Don't be afraid to admit that to yourself.

Once you're confident with your results and reach that elusive burn rate, move on to another page element and start the process again. Search engine optimization isn't about perfection; it's about balancing your "good enoughs".

But don't forget to come back in the future. Once you think you have a page optimized, it doesn't mean you'll have it optimized forever. Your users and what they're looking for change so frequently that it's important to go back and restart old tests. You can step away, but you have to come back and retest and re-measure things like clicks, engagement indicators, user behavior, etc.

Posted by Lisa Barone on 06/19/08 at 2:42 PM | Comments (0)
See more entries in Analytics

June 4, 2008

Analytics Every SEO Needs To Know

I'm stepping in late after a lunchtime interview for SEM Synergy with the awesome Jeremiah Andrick. Check that out next Wednesday on WebmasterRadio.fm.

On to the session. We've got Rand Fishkin, Co-Founder and CEO, SEOmoz, Inc., moderating. Our speakers are Brian Klais, Executive Vice President, Search, Netconcepts; Laura Lippay, Group Program Manager, Search Strategy, Yahoo!; Jonah Stein, Founder, ItsTheROI; and Richard Zwicky, President, Enquisite.

Brian Klais is up first and is going over a list of metrics that you need to know. Non-yielding pages should be identified and know the rate that each yielding page is generating. Look for inequality in page placement. Measure the engine yield rate. ROI and brand reach are critical metrics. The missed opportunity cost can be calculated, based on assumptions that all your unique pages should be yielding, and that the average number of visitors per page apply to all pages. For more on the subject, search for "natural search KPIs".

Laura Lippay is going to share how to prove the worth of SEO. She says that the Grid is the way to show what you've accomplished. It's a collection of keyword-based data, stored in an Excel file or database, which can be pulled into multiple varied reports.

What can you do with it?

• Balance SEO, PPC, and Paid Inclusion (PI)
• Find SEO referral gaps
• Find SEO content opportunities
• Make SEO traffic and value projections

What do you need?

• Keywords
• Number of Searches
• Conversion metric ($ per PV, LTV, etc.)
• PPC data
• PI
• Algo
• Search engine CTR by position

What to use it for?

1. Gather keywords (and optionally, the number of searches)
2. Add keyword data for performance comparisons
3. Or just to see performance of one channel
4. With search volume and CTR by position, make assumptions

Once you have the data, showcase your skills in the grid. Anything keyword based can be part of this grid.

Jonah Stein says he's going to talk about the five forgotten metrics:

1. Customer lifetime value
2. Crawl frequency - the new page
3. Page views to conversion
4. External links
5. Webmaster Central all query stats

To manage your customer relationships, write a permanent cookie on First Touch, write First Touch data to CRM on conversion event, and capture missing data at every touch point. With that data you get more accurate LTV & ROI.

Crawl frequency is one of the most meaningful metrics. CF is a product of search engine's score of your page. Obfuscating CF lowers index quality. Relative CF is great for diagnostics. Crawl frequency is governed by the number and quality of inbound links, the content update frequency, server performance and Sitemap settings.

Metrics for crawl frequency include:

• How often crawler visits
• Crawl depth
• Crawl saturation
• Crawl frequency rank by page (infrequent, supplemental)
• Pages that don't get crawled have issues

How to measure crawl frequency:

• Crawl rate tracker
• SEOmeter.com
• Log file analysis
• Develop a custom solution

The little-used metric of page views to conversion will illuminate that good SEO gets the user closer to conversion. Decreased page views can be good or bad. Decreased time on page can be good or bad. And decreased time on a site can be good or bad. To measure this, you need to know the funnel for each landing page and to measure success to the funnel. As for external links, measure the total number of external links and the external links per page.

Webmaster Central for all query stats should be looked at. Sometimes the content is not human readable. A converter for Google WMC stats can help you decipher the data. The keyword ranking is shown by directory. CMS developers and large site publishers should integrate into the back end by importing WMC query stats, importing external linking data, and displaying internal linking data.

Richard Zwicky is next, and will somewhat base his presentation on a recent blog post he published. Using SEOmoz as an example, he's walking through their analytics data on Enquisite software. The program allows you to look at the location queries are coming from and look at multiple keywords. Once he knows what terms are bringing people to a page, and look at what's not performing in your target markets.

Q&A

What tips or best practices do you have for measuring 2.0 data?

Jonah says that you can tag AJAX states to get a map of what's going on and create an alternate value. That way you can tag what user has completed what action.

Laura, what measure do you use to measure the lift between different search engines?

She says she bases it on models that are out there, available online. She doesn't get search data just because she's at Yahoo. She has access to what everyone else does.

When performing human evaluation of the data, is it meaningful to look at the top 200, 1,000?

Brian says that you have to position yourself implement scalable solutions. It's kind of looking at like business metrics, it's a signal to look at the health of the business. Richard says that the tail end is what needs to be focused on. Jonah says that if your site has a hierarchical structure, just look at the landing pages that delve deeper into your site. It's easier than looking at a list of 200,000 pages.

Does the panel have any recommendations for log file analysis to track spider behavior if you don't have the resources to custom build?

Marty Weintraub, sitting in the audience says that ClickTracks has an API for log file. However that still requires custom development. Someone else in the audience says ClickTracks Pro can do it. Brian says some clients use programs like Omniture or Hitwise.

Is Omniture worth the cost?

Laura says it's always worth it. Richard says some people run their whole company around it. He says that some customers fall back on privacy of data. There's more sampling and more opportunities to slice and dice your data on Omniture. Jonah says the threshold is whether or not there's a full-time analyst looking at the data all day. If so, then it's probably worth it. If it's only looked at for two hours a month, then maybe it's not worth the cost.

Posted by Virginia Nussey on 06/ 4/08 at 3:10 PM | Comments (3)
See more entries in Analytics, Liveblog, SMX Advanced 2008

What You Should Be Measuring -- But Aren't

Time for some more paid search! Chris Sherman is moderating with speakers Akin Arikan (Unica Corporation), Christine Churchill (KeyRelevance), Rich Devine (ZAAZ), and Ryan Gibson (Rimm-Kaufman Group).

No witty banter this time as Chris gets right into things. Christine Churchill is up.

"Sixty-three percent of consumers who conducted online searches for various product categories convert offline" - Google, comScore study March 2006

Many companies push prospects to a phone call or other offline channel because it allows marketers to use the power of voice to convert prospects. This tactic is appropriate for businesses requiring "high tough". It works for B2B and companies with long complicated sales processes, and is common for local service companies.

Methods to Track Offline Conversions

Simple:

• Make an assumption based on their sales
• Anecdotal data: Ask the salesperson or call center people to ask customers how they learned where to buy the item.
• In Store surveys: Surveys ask customers how they found the product.

Intermediate:

• SWAG it from limited pilot tests
• Use a unique phone number in ads from pilot campaign
• Determine an online-offline ratio
• Extrapolate future sales
• Offer coupons or special offer codes
• Unique pricing: Place unique online pricing on a search landing page. If people quote the special rate then you know they started online.

Advanced:

• Customer tagging: Tie the online cookie with the offline customer number or credit card data of offline purchases.
• Use of Unique Phone Numbers: Users from different referrers provided a unique landing page with a unique 800#. The Customer is with a cookie so the phone number is used consistently.
• JavaScript overrides different number based on referrer.
• Phone per campaign
• Track via phone logs
• Pay Per Call - Track calls on separate ad networks

Points to Remember:

Many people do research online and buy offline
Normal online analytics don't count offline conversions
Offline conversions can be considerable depending on your business
Offline conversion information is vital to making informed online marketing decisions.

You need the whole picture to make good business decisions.

Ryan Gibson is up.

Ryan says you have to start off by making some assumptions: That you're managing at least to keyword level-preferably ad level, that the ultimate program goal is to drive profit, and that there's a difference between brand and non-branded terms.

Typical Measurements:

• A/S: Ad Spend to sales
• ROAS: Return on Ad Spend
• ROI: Return on Investment
• CPO/CPA: Cost per Order/Acquisition

You want to look at these metrics on the keyword level and look for real actions happening on the Web site. When you're looking at some of these metrics, the proxies are all based on margins. Typically you see that the folks with the largest margin are going to have the highest spend targets.

Ryan throws out a really gross looking formula:

A/S = (1 - COGS [cost of goods sold]- Variable Costs)/2
ROAS = 1/(A/S)

That equal typically targets roughly ½ of the margin, so why not use the margin directly?

Why is margin so essential?

There are different margins on different product categories in your marketing mix. Accessories will have a higher margin value than the main product.

Existing Tracking: Data can be passed to your bidding system in addition to the sales of dollars.
Post Sales Data Feed: Nightly process to match up orders to the margins.
The Best: Feeding the margin data on the fly coupled with a nightly process to knock out frauds and cancels.

You still need an understanding of the lifetime value. What other actions are of value on the site: email sign up, request a catalog, other downloads, etc.

Well, that was confusing. Mostly for me. I'm sure the other, smarter people got it.

Up now is Rich Devine. My toes are cold.

Conversions don't tell the whole story. When that's all you're focusing on money is left on the table and you miss the big picture. Not assigning full credit to marketing efforts.

This is related to hot topics like engagement, attribution, keyword stacking, etc.

He talks about building a custom-built performance model that assigns dollar values to key site behaviors in order to understand express the value of your Web channel at large. It's really useful for lead generation. Key site behavior include: locating an online dealer, viewing a product showcase, orders through a call center, etc.

Building a Monetization Model

• Confirm business goals
• Identify and align site goals to business goals
• Establish an accurate, key site metrics
• Id key site behaviors or micro conversions
• Assign value to key site behaviors
• Then...have a beer (or a Sprite).
• Go to town with monetization.

Tips for Building a Monetization Model

Discover and use all available data sources. Stuff like Web analytics, CRM data, financial data, etc. Anything you think that will contribute to the model.

[He's zipping through these slides like he's hungry for a Sprite. My attempt at keeping up = teh fail.]

Prioritization of optimization activities: Conversion is not the primary decision driver. Are you assigning too much value to the converting keywords? Are you undervaluing non conversion keywords?

[Missing. Everything]

Tips on Getting Starting with a Monetization Model: It's a model - don't freak; It's not perfect. It's okay to have a wild ass guess as long as it's a scientific wild ass guess. Be comprehensive in your data sources. Get quantitative and qualitative. Be accurate but don't get too strict.

Akin Arikan is next.

Is paid search happening in a vacuum? What about the blogosphere? Do they impact your search? What about everything else?

Search is not alone. There are online ads, offline ads, direct marketing, social media, and relationship marketing.

Case Study From Covario

They did time sequence testing where online ads that were running in parallel to search were turn off. The results:

• Click through rates when down 363 percent
• Conversion rates when down 33 percent
• Had to pay 16 percent more per click to maintain the same business volume

Do control groups by geographic targeting. This applies to offline as well. If you plaster ads all over the streets of Seattle, do you get a surge in traffic?

Search vs Web 2.0 Participation

He talks about UPS Whiteboard viral campaign. I haven't heard of it but maybe you have. If you're UPS, you own this microsite so you can set a cookie for users that visit to see what they're doing afterwards. For that to work, when they go to coolwhiteboards.com it should redirect them to whiteboards.ups.com.

He brings up an engagement funnel that records how many participated, referred a friend, etc.

What about the power of the blogosphere?

People who heard about a bad shopping experience are less likely than the people who actually had the bad experience to ever set foot in the store. [Most. Awesome. Stat. Evar.]

There is blog monitoring technology that does text mining to get to a sentiment of what's happening. They also do trends to look at chatter and effects.

Relationship Marketing: Mary receives a catalog. Two weeks later she visits the Web site via the search. Did the catalog cause the visit? You can do Matchback. You know who you sent the catalog and if Mary is a registered user on your Web site than you can 'match it back'.

It's time to take off the blinders. Talk to the guys on the online/offline/brand marketing site. We have a lot to learn from each other. It's not just technology.

Question and Answer

What method do you use to find the value of your micro conversions?

Rich: The first thing is to get what you can out of the analytics. Identify 4 or 5 core site behaviors and focus on what your core business objective is.

Christine: Identify what your measurement of success is. She takes the micro conversions as some of those measurements of success. They're going to be different for everyone.

Ryan: Ask if those micro conversions are happening offline. What are they costing you offline? What's the value online? Compare them?

What keyword should get the credit for the sale, the first or the last?

Akin: You want to make specific studies at some point to get a real answer. Set up some control groups. You can't do that at all time. At some point you need to reach a conclusion and then create a rule. The details of what you pick may result on your bus model and case study.

Ryan: They've found that a lot of multiple clicks that occur in paid search happen on the same keyword.

Christine: It's going to vary on who your company is. If you're a big brand you don't show up for [movie], you have a problem. But if you're an independent film maker, a long tail term may be more appropriate.

How would you a handle a sit where clients set conflicting metrics to evaluate success?

Rich: They try and address that before they execute. Sometimes it takes a long time, especially with bigger enterprise client. But before they begin to execute they come to some sort of agreement and make sure that everyone signs off. It helps to relieve a lot of the headache.

Christine: One of the first conversions you want to have with a client is about what they're measurements of success are. You want to set that up upfront. Managing expectations is key. Clients aren't stupid. They may be seeing things from a different viewpoint so keeping the communication open is vital.

Posted by Lisa Barone on 06/ 4/08 at 12:16 PM | Comments (0)
See more entries in Analytics, Liveblog, Pay Per Click / Online Ads, SMX Advanced 2008

June 3, 2008

Closing the Loop: Are You Tracking Every Lead?

Chris Sherman, Executive Editor, Search Engine Land, is moderating. Our speakers are Adam Goldberg, Co-Founder and Chief Innovation Officer, ClearSaleing; Alissa Ruehl, Manager of Paid Search Services, Apogee Search; John Tawadros, Chief Operating Officer, iProspect; and Lauren Vaccarello, Director of SEM and Analytics, FXCM.

John Tawadros starts off. What if you could tell what your leads are doing throughout the day? He's talking about tracking leads and how much you should track.

What can be done with the information? Search converts 21 percent better if a user is exposed to a banner ad first. 37 percent of the time users look in at least two engines at any given time. There is no impact to conversion beyond 6 touch points. Print ads improve CTR by 33 percent. He's giving us a lot of stats.

Does it make sense? Is data size statistically significant? Expectations - What if you learn nothing monumental? Cal you live with the imperfections? Can you drive every channel through one tracking system? Can you support the investment?

Chris asks if there is a rule of thumb: Is there too much you can track. He says it depends on how much you want to spend, but for him, there's never too much because you just don't know what you don't know.

Alissa Ruehl is talking about tracking past the lead and into the sales funnel. What does your company really want out of your marketing efforts? Is it sales? Top line revenue? Bottom line revenue? ROI? What metrics are you currently using to measure your success?

All leads are not created equal, so how can you separate the good ones from the bad? Integrate your paid search system with CRM. Pull paid search info into your CRM system. Report on sales instead of just leads.

She's got three case studies.

1. Client 1: ROAS by keyword
   • They found that keywords could generate leads, but not sales or revenue, and it went the other way around, too. Hidden treasures and shallow successes were outed.
2. Client 2: ROAS by engine
   • Yahoo was not generating leads as well as Google. Looking at the CRM system, they found that the cost per sale in Google was over $750, while for Yahoo was under $350. This changed the client's first intention to cut back on Yahoo spend.
3. Client 3: ROAS by Web site
   • The new design had a slightly higher click to lead conversion rate. However, the new design actually had nearly double the cost per sale.

Research of data was needed to uncover what superficially looked like a success but was really a loss, and vice versa.

How is this done?

1. Create custom fields in your CRM system for the info you want to capture, for example, keyword, engine, campaign, referring URL.
2. Update tracking URLs. If you're already capturing this info in your analytics system, you can use the same variables.
3. Set cookies to capture these variables. Put it on all pages, not just landing pages.
4. Add hidden fields for each of these variables to all of your forms. These should pull values from the cookie rather than user input.
5. Test and report. Make sure the process is working. You should see leads coming through with paid search tracking info, as well as basic contact info from the form. Set up reports in your CRM system to list sales or opportunities.
6. Compare this data to your paid search spend, just as you do with your lead data.

Caveats to remember:

• Pay attention to statistical significance.
• Take your sales cycle length into account.
• Look at junk leads as well as sales.
• Consider an intermediate step of looking at cost per opportunity.

Adam Goldberg agrees that all leads are not created equal. In fact, that's what he'll be talking about. He says that if the cost per lead is too high, it's considered bad. If it's too low, then it may be because of high leads per sale. But this thinking can be very misleading because the high costing lead could be the highest profit generator.

Next he's using a basketball metaphor. This will be interesting. Points = sales and shooting percentage = conversion rate. If you take out the player with less points, you may be taking out the person who's providing valuable assists.

Stages of the customer buying cycle:

1. Problem recognition: banner recognition and emails
2. Information search: your paid search ad
3. Evaluation of alternatives: they look at your competitors or other options
4. Purchase decisions: they go back to the search engine with the specific brand
5. Purchase: cha-ching

You could think of allocating your profit equally among all the steps OR you could put more money into the steps that put the product into the customer's mind, not step 4 where they already know what they're going to get.

He has a slide of an advertising ecosystem. At the top is Advertising Sources, pointing down to Capture Advertising Analytics. This points down to four options: No Sale, Call, Submit Lead Form, and Purchase. Both Call and Submit Lead Form point to CRM Software.

Lauren Vaccarello is up to the podium next. She says that people working in e-commerce don't really need to listen because they get the data from shopping carts, whereas lead gen people should listen.

The benefits of integration:

• Bridging the gap between sales and marketing
• Remedy the size fits all conversion tracking
• True conversion tracking

You have several options for integration:

• Omniture
• HBX (Visual Sciences via Omniture)
• Clicktracks via Hotbanana
• WebTrends
• Google Adwords
• Custom programming with your solution or Google Analytics

One potential obstacle is sales objections. The solution is bribery. Give the better lead lists to those who are on board.

Another potential issue is poor planning. The solution is do your research first, set your goals before you start, and define and add fields before integration.

Strategies to increase ROI:

• Go after the low hanging fruit. Build campaigns with highest conversion rate reports.
• Build Time to Sale reports.
• Create marketing campaigns geared toward long sales cycle leads.
• Get your highest ROI by campaign report.
• Stop wasting money.

Takeaways:

• CRM integration is not for everyone.
• With integration you can garner true conversion tracking.
• You have a variety of integrating options.
• Do your homework first.
• Prioritize your leads.

Posted by Virginia Nussey on 06/ 3/08 at 4:33 PM | Comments (0)
See more entries in Analytics, Liveblog, SEM Events, SMX Advanced 2008

May 27, 2008

SEO Weekend Update

Hi, Friends. I hope you all enjoyed your long weekend. Because now it's over and it's time to get back to work. IT also replaced my keyboard this morning which has completely confused my fingers. For every five words typed, I'm almost guaranteed to have one spelled correctly. Huzzah for productivity!

One Week 'Til SMX Advanced!

I checked a calendar earlier and was shocked to realize that SMX Advanced is just one week away! If you waited until the last minute to purchase an All Access Pass, you're now officially out of luck because they're sold out! However, you can still purchase a Developer Day, Networking or Expo Only pass before the big show. You better hurry.

Of course we'll be sending a Bruce Clay contingent to man the booth and take photos with babies provide demos of our SEOToolSet. Bruce will also be speaking on the Creating Value In Your SEM Businesses panel and we'll be providing a double dose of liveblogging coverage of the event.

SMX Advanced will also be your first chance to meet two new Bruce Clay, Inc. employees. Our Associate Writer Virginia Nussey will help me on the liveblogging front while recording some new SEM Synergy episodes. You'll also get to meet Chris Hart, the face running our new East Coast office! Chris will help Bruce Clay rock the Long Island pride (and patented accent) and help us bring our SEO training program to the other side of the States. We couldn't be more excited to have him on board or to announce the new office. Look for a press release early next week for more information about Bruce Clay, Inc. East!

Why Do You Put Up With Twitter?

Twitter continues to be as unreliable as our posting schedule was back when Susan ran the blog, so why are you still using it? Loren Feldman says it all goes back to our need for attention and to feel that someone is listening to us. Unlike your blog, on Twitter you can offer up a 50 character message about your lunch and people will immediately respond and validate your existence. Is that all Twitter really is? Or is just habit at this point? I mean, if you left Twitter, for say, Friendfeed, you'd have to go through the whole "setting it up" hassle again. Who wants to do that?

It's interesting, though. Because for as many posts as we see circulating Sphinn and beyond all talking about how unreliable Twitter is, we're all still using it. [Often to complain about how annoying it is.--Susan] Why are you still staying Twitter-loyal? Listen to what Loren has to say and then offer up your own two cents.

Best Practices Aren't Always Best

Nathania Johnson over at Search Engine Watch pointed me to a great post on the Google Website Optimizer blog that proves that best practices aren't law. The post discusses a landing page case study that was done for a stair remodeling company. Two landing pages were tested -- The "pretty" one that, by all accounts, should have come out on top and a landing page without the model or the visible menu. At the end of the test, it was revealed that the "inferior" page actually increased the company's conversion rate by 144 percent and grew the average order size by 18 percent.

I'm not surprised. Having attended Jim Sterne's eMetrics Summit earlier this month, that lesson was drilled into attendee's heads. So many times marketers assume they know what's best or what a customer wants when they couldn't be more wrong.

The lesson of the day: Test everything. Even best practices can be wrong.

Fun Finds

ABI Research says that the Internet video audience will hit one billion viewers by the year 2013.

Mashable talks about the rumored Social Bookmarks site Microsoft is set to be releasing.

Posted by Lisa Barone on 05/27/08 at 2:24 PM | Comments (0)
See more entries in Analytics, SEM Events

May 7, 2008

Measuring the Success of Non-Commerce Websites

I just had some coffee. I feel better.

Alex Langshur (PublicInsite) is here to talk about how to measure success when it's not about sales and shopping carts. I'll admit; I've been looking forward to this one since I got here. I hope it delivers.

Alex jokes that his session is up against Avinash Kaushik and Jacob Nielsen so he has a bone to pick with Jim Sterne. Heh.

Alex starts off saying that there are plenty of institutions where dollars are not part of the equation. They're working a lot with Harvard University. Harvard is struggling to find the value of the online channel to their organization and they're trying to help them with that.

Every time they try and quantify the value of the online channel people are still looking at shopping carts. It's like trying to shove a square peg into a round hole. What's interesting is that sometimes you can actually use a square peg to fit the round hole problem. Like with square watermelons!

Engagement: It's a buzzword but we're stuck with it. If you go and look at Eric Peterson's engagement equation, it's still a number. And numbers don't necessarily help you manage all that well. It's one way of approaching it but he's not going to focus on it.

What should I measure?

When you try and reduce really complicated sites that are an expression of the organization into a simple number, it doesn't work. You have to find another way. There are two things you need to look at:

Pre-click behaviors: These are leading indicators like visibility and alignment
Post-click behaviors: These are lagging indicators like conversions and intensity.

Pre-click is going to drive post-click behavior. You have to use benchmarks in order to give these things value. If you're not using benchmarks because there's "nobody like you", you're missing it.

Benchmarks are really important because they give you perspective. It helps you to understand potential audience size and where you sit relevant to others. Excellent free sources include Google Analytics, Compete.com, Quantcast,com, index.fireclicks.com and your own analytics vendor.

Measurework framework:

Visibility: What affects our online profile? SEO (recommends using Yahoo Site Explorer to check backlinks), traffic acquisition campaigns, buzz monitoring (he recommends Trackur). Who's talking about you, where, when & what is being said?

Alignment: What is our alignment to the audience and its needs? Look at the intent. What keyword themes bring visitors to the site? Look at your search analytics. Also pay attention to audience segmentation. Are we reaching the right audience? Are we tracking the right audience? Can we segment, and if so, how?

When you're doing keyword research, there are a million of opportunities for segmentation and alignment in the long tail. The head may be your bread, but the long tail is your issues. They're the authors on your Web site, the books you're publishing etc. You need to look into it.

You do that by reverse engineering the Information Architecture and breaking the long tail terms into categories. You can use text mining tools to help you, but you really have to create the categories and themes by hand.

Another part of alignment means asking yourself if you're getting the right people. Do these 1500 visits you're getting from Europe influence the key site metrics? If they're not aligned to you, those visitors degrade your stats and skew your metrics in a negative way. You must find a way to address that. It may mean playing with your geo-targeting.

Intensity: What is the change in key performance indicators for visits? There are a wide range of KPIs to choose from. Things like page views/visits, average visitor duration, unique visitors, user generated content volume, etc. Don't track them but all figure out which ones matter to you. The Web Analytics Association has a big book of KPIs. You can reference that. If you go beyond 15 KPIs, you have too many. He prefers to go down to 10. When you're dealing with non-commerce based organizations you have to crystallize the mission or you risk losing the discussion when people become too tied up with the numbers. Don't design the indicators by committee.

Conversions: How are value events performing? Things like email sign ups, RSS feeds and clickthroughs, downloads of key content, AJAX/Flash events, Send to a friend/print pages, social tags and bookmarking, and satisfaction survey results. In a Web 2.0 world you can use tools like Feedburner to track value events.

How should I report?

• Don't provide access to raw data, reports or the tool
• Find the hook that can start a conversation, then build the story
• Take the time to report less and inform more
• Contextualize data in terms that matter to the recipient
• Look for champions in the organization
• Nothing succeeds like success

Case Study: Public Sector, health promotion

Goals: Encourage visitors to inquire about healthy eating and inform them of the benefits of healthy eating

Desired Outcome: Encourage them to take action that reflects a shift in their thinking about that subject.

Visibility: Find actions that would raise awareness - SEO
Alignment: What is KW analysis saying about intent?
Intensity: How are they using the site?
Conversions: What are the outcomes?

Visibility: If they hadn't raised visibility, they would have had a very low amount of traffic. It was one of those things where you can do great stuff, but if you can't get people to come to the site, no one will know about it. And if that happens, you won't get the money to do it again in the future.

Alignment: They were looking for people who were at home and people in the K-12 segment. They had 20 percent of their audience from those segments. It was a lot higher than previously.

Intensity: People were coming but they weren't coming back. Wasn't a lot of intensity of use. That was a problem going forward.

Conversions: Terrible conversion ratio of .26...but it was double last years. They had a fairly respectable sign up rate. How do you report that?

Created a color-coded scorecard that explained the define objective, the result and a color-code that said if the objective was met. The purpose was to start the conversation. They knew if they gave them a report no one would read it. They didn't send it by email. They printed it and distributed it by hand.

Question & Answer

Do you have any preferred methods for doing long tail keyword analysis and putting things into buckets?

The short answer is no. They've run studies looking at the top 100, 200, 300, 400, etc keywords. The curve is like a normal distribution where once you go beyond 600 keywords, the deltas are under a percent. If you do the Top 500 you're going to be capturing the major portion of what the intent is. If you still to that, you'll have a good understanding as to what's going on.

You talk about profiling segments. It's hard to do in practice. How have you gone about profiling segments?

1. Geographically profiling: You HAVE to be doing this.
2. Profile-based navigation
3. Qualitative analysis

How do you establish ROI?

In a public sector org, ROI is meaningless because you can't bring it back to profit and loss centers. In a nonprofit org, it might be a little bit easier but it's tough to do. He believes in establishing ROI based on programs and initiations. ROI is "are you reaching the right target audience?" He goes back to those kinds of value adds.

Posted by Lisa Barone on 05/ 7/08 at 12:35 PM | Comments (0)
See more entries in Analytics, Liveblog, eMetrics San Francisco 2008