As marketers, we spend a ton of time implementing stuff — but what are the important things to measure, and how do we measure them? That’s what we’re here to talk about in our first liveblog session on Day 3 of PubCon.

First up is Alan K’necht. He asks, what are we measuring now?

First, this is a big mistake:

• Heads, fans, etc. Are they engaged? Is anyone interacting with them?
• Paces: number of posts, updates, tweets (these are milestones only)

We should be measuring victory and success.

Tools of today (not recommendations, just some useful tools); the biggest mistake is when people don’t use these tools as intended.

• Klout
• Twittalyzer
• PeerIndex
• Webtrends

Measuring fuel:

• Fans, followers, etc. We can measure quantity and quality.
• Scoring tools: (see above). How big is your pond? If you’re using a scoring tool, does the size of the fish matter or is it the size of the pond?
• Are you growing your following base? Are they engaged?

Measuring oxygen (the people):

• Brand advocates (high Klout scores, for example). Use tools to help you find the right fish in the pond.

Measuring the spark/heat:

•  This is what is going to get people talking about you; good content. This gets the fire growing.

Measuring fire:

• Analytics uses ratios; which content is engaging, which gets spread? Likes and content shares per post, for example.
• Biggest mistake: posting links within Facebook and Twitter. Use URL shorteners in Twitter or else you won’t be able to track the link in analytics. Use the campaign code in it.

Bring it together:

• Look at increase in references; more talk about the brand (he uses radian6 for some clients).
• Is there a correlation in sales?

Measure attraction, not BTUs; Draw people in like a campfire. Be there with them and make it successful.

Up next is Tom Critchlow. He starts off by saying data is a marketing asset. He is saying to read the OK Cupid blog which has a ton of data, like sexual activity by phone brand. They gather the data and use it for marketing. You can’t get this data elsewhere; they are using as a technique for marketing.

Business Insider actually publishes its live analytics online for a marketing tool that anyone can see.

If you’re going to use data in marketing, he urges you to use data visualization instead of infographics. Data visualization is where infographics are headed. Bad infographics will die out.

There’s very few metrics you need to care about, and the ones you do need to care about, you should watch like a hawk. It’s not just a set of metrics you watch, they must be the key drivers of your success.

Measure things you have direct control of. Rather than measuring the number of people that come to your site, measure the pieces of content you’re doing. Measure the activity as well as the results.

Data should drive actions. If you want to get insight out of data, you should form hypotheses first and then use analytics to test that hypothesis. It’s less efficient that just wading through data.

The SEO community is heavily driven by page quality. He is showing an example of a page that has a lot of products, and one product has reviews compared to a similar product page with no reviews. How do you tell the difference between this product page in analytics? Use customer variables to help. You can also measure the number of people coming to your site that are Facebook fans by just using JavaScript.

User feedback is data, too. As marketers, we want to convince people and influence change but we don’t have great data to back it up. You can survey people and measure that way if you don’t have the number data.

As social media continues to grow in popularity and more time and money is invested into it, pages continue to be placed on Facebook with little thought into analytics and measurement.

With the death of the Static FBML page and the ability to now use iFrame’s to display tabs, analytics software – such as Google Analytics – can now be used to track visits, engagement and – where tabs include specific tasks – conversions.

At Bruce Clay Australia you’ll notice there’s been a lot of talk around conversion rate optimisation. This thinking shouldn’t stop at your website, but should be carried over into other areas such as Facebook. You want to know that when people hit your Facebook page, they are being given the greatest incentive to convert as possible – be that hitting the ‘Like’ button, watching a video or signing up to a newsletter.

So what can, and should, you be tracking on your Facebook page?

As with anything the answer to this will largely depend on the content of the tab and the goals your business has for Facebook. This post will aim to give you a “first steps” guide to getting everything in place, and will also cover some topics you should be thinking about before pushing tabs live.

Adding Google Analytics to Tabs:

This is the easy bit, but will depend on what kind of tab you’re using.

The recommended approach is to develop your own Canvas App, which can be done here on the Facebook Developers site.

The benefits of developing your own Canvas App include being able to customise the logo used in navigation and ensuring the iFrame is hosted on a server which is big and fast enough to handle perceived bandwidth; particularly if you’re running a promotion and are expecting large volumes of traffic. The last thing you want is the tab to crash because it can’t handle the traffic.

I’ve found the standard Asynchronous GA script works fine with this setup as long as the URL that is associated with the iFrame is the same as the URL associated with the primary Google Analytics site.

So if your Google Analytics account sits on the site:

• http://www.mysite.com/

and the Facebook tab sits on a URL such as:

• http://www.mysite.com/facebook-tab/

The code you’ll be using would be:

As with adding Google Analytics to normal sites, the above code would be placed just before the < /head > tag, and you would change the UA-12345-1 number to the one associated with your GA account.

If you’re not creating an app of your own and are using a free tool such as Static HTML: iframe tabs, or if your iFrame is referencing a different domain than the one associated with your Google Analytics account, you’ll need to add two extra lines to the GA script to track between multiple domains.

The new lines you’ll be adding are:

And they fit into the GA script as per the bolded lines below:

If you’re linking between the Facebook page and your website it can also be helpful to add the _link function to any links between the two. This is gone over in greater on Cross-Domain Tracking page on the Google Code site, however in a nutshell it involves adding the following code to links:

The new version of analytics (v5) became available to all users from April 2011. Other than its new look, it has many new features for helping provide better analysis and SEO reports. The analytics team is still working to give us more surprises over the next few months. I will be covering the most useful features that were missing in the old version of analytics.

Simplified Custom Reports

The Custom Reports section of the new analytics has become more advanced and manageable compared to the older version. Now users can create new categories that group the related Custom Reports simply by clicking a few buttons. There is also an option to migrate all custom reports from the old version of analytics.

Another improvement is the ability to quickly re-order metrics and dimensions using drag and drop feature. All these simple features help to create reports the way we want in short amount of time. To learn how to create custom reports in the new Google Analytics, visit the custom reports article in the Google Analytics Blog.

The new Map Overlay reports

In the new version of analytics, we can now see region level maps of over 170 countries. The previous version had the map overlay feature within the visitors > demographics report but now it is available in other sections of report. This feature can help provide a broader view for analysing geographical data.

New map overlay feature (Source: Google Analytics Blog)

To try this feature, simply login to Analytics, select the new version if you have not already done so. Go to the Map Overlay report that is under Visitors > Mobile > Devices and then click on any country. To see cities, simply click on the City and in this view, you might be able to see a magnifying glass that appears when cities are clustered closely together just like in the image above.

New metrics and dimensions

The new metrics are always better because we get to do deeper analysis of our website using many choices. The analytics team added 45 new metrics and dimensions in the custom reports. The custom reports now can also be explored without making changes to the actual report just by selecting secondary dimensions in the explorer tab.

Categorised metrics and dimensions (Source: Google Analytics Blog)

The new metrics and dimensions cover more variety of areas such as social, e-commerce, mobile and advertising. To make it easier to use, all the dimensions and metrics are now categorised.

New categories

Another new feature of Google Analytics v5 is the new categories that can be found in the left navigation. The most useful including:

• Social reports: Once you have implemented the tracking code on your website, you will be able to measure the impact of social interactions on your websites. The social engagement report shows behaviour of your site for visits that include clicks on sharing buttons like Follow, Like and +1.The Social actions report shows the amount of social media activity on the site (Tweets,+1’s, Facebook Likes etc).Social pages report shows the pages on the site that drives the most social actions.

• Site Speed: Site speed has become an important factor in SEO, and the new version of analytics now reports on the site speed. This report can be useful to answer questions like: Which landing pages are the slowest? Or how does page load speed vary across geographies?

To setup the site speed report, you have to enable it in the analytics and update your tracking code. The detailed instruction on configuring and using site speed can be found here.

I hope these new features will help you do better analytics and create good reports. Stay tuned to Bruce Clay Australia’s blog for more Internet Marketing news

Moderator: Richard Zwicky, Independent Consultant,

Speakers:
Ray “Catfish” Comstock, Director of SEO, BusinessOnLine
Todd Friesen, Director of SEO, Performics
Rob Garner, Vice President Strategy, iCrossing

This has been a popular session for about the last 6 years and this year’s panel looks to be a great group.  Rob Garner will speak first today, then Todd Friesen and lastly will be Ray “Catfish” Comstock.

Rob will start us off today by talking about measuring SEO success.  He shows a slide that says, “Paid vs. SEO: the 90/20 search budget conundrum”.  This basically says that 90% of the search channel budgets go to paid search while paid search only gives 20% returns.  On the flip side, 10% of budgets go to SEO and those playing in the organic space will usually see about an 80% return on all revenue.

Why is it so upside down, Rob asks. Because paid is easier to track and measure. With SEO it’s harder to look longer term and tie direct ROI back to specific engagements. However, Rob tells us that SEO can be easy to measure in aggregate and at the keyword level once a balanced program is in place. He states that the biggest opportunities for marketing in the search channel are in natural search.

Some of the challenges of we all face with tracking SEO are:

• Disconnect between legacy systems in tracking performance
• Using last-click attribution
• Larger companies measure value as “what have you done for me lately” – natural search benefits tend to be longer term
• Not maintaining search performance history going back years
  • You can’t compare what you did last month to last year…but instead stack it and look at things like seasonality but as a whole
• Bad analytics and tracking
  • Analytics utilize algorithms too – make sure yours is sophisticated in the search tracking space
  • You need a program that gives you a robust look at things. It needs to bucket things in meaningful ways, show you long tail keywords etc.
• Educating those in charge of the budgets for natural search
  • Constantly educate those influencers who hold the money

Along with the challenges of tracking SEO success, we are often faced with some tough questions that usually do not have simple answers.  Questions like ‘how much is domain authority and trust worth?’ or ‘How much is a link worth?’, ‘How much is a link development campaign going to return on investment?’ and ‘What is the ROI on implementing the meta description tags?’.  There usually is no logical answer that the higher ups will accept, so Rob suggests just trying to avoid answering them if possible.

With SEO, a longer term view is needed due to the fact that natural traffic is earned and eventually owned so it becomes a traffic generating asset. On top of that, the maintenance costs for SEO are a realistic expense.

4 key measures needed for meaningful SEO metrics are:

• Value of search equity for the site being optimized
• Media value of natural search
• Value of various action across the site
  • Both individually and in aggregate
• Value of the stress and time costs on your organization
  • You can do it now 10x easier than if you do it later

If you can put a number to those 4 things, then you will be able to measure how important and successful a SEO campaign can and will be.

If you’re having trouble convincing those holding the purse, then here are some tips to help you build a business case. Find out the following:

• How much is invested in search overall
  • And in particular, how much is spent on SEO activities
• How much is a conversion worth and how much will search modifications be worth
  • What values do we place on various actions
  • This is a good opportunity to look at words performing well in paid, but have no rankings in natural to see if you can capitalize on it
• How do we measure lift in natural search metrics
  • Project traffic lifts based on increase in content and relevant theme of pages against keyword search stats
  • Link dev – measure lift in rising keyword ranking against traffic, against conversion
  • Create a natural search event
• How much opportunity may be lost without an SEO plan
  • What is the value of current traffic, and how will site redesign impact traffic

An example Rob gives is about establishing the market value of a search click:

• Calculate anticipated traffic lost or gained in natural search
• Determine the average CPC in a themed area of one of your campaigns
  • This is the marked value that a marketer is willing to spend on relevant traffic based on actual spend
• Multiply the average CPC times the amount of SEO traffic to quantify its media value
• For example: 100,000 additional clicks in natural search per month times an average CPC of $2.53 is valued at $253,000. In a year, $3,036,000 in media value of traffic

Is your site going through a redesign soon?  Here are some potential issues that can come out of a redesign:

• Loss in natural traffic
• Loss in link equity
• Loss in natural search equity and history
• Costs for fixing technical mistakes
• Cost for recreating containers for content
• Loss in sales, actual monetary returns
• Loss in trophy rankings
• Loss of the sum of long tail rankings

Make sure any redesign has a redirection plan (301 redirects) so that your site doesn’t lose traffic!  This can hurt the site in more ways than one, especially if there isn’t a custom error page in place.

Overall, marketers must place a monetary value on actions to help SEOs quantify results. Marketers spend lots of money on media, but they don’t often place a value on actions. They also don’t place an educated value on actions from natural search (or any media spend) essentially spending blindly. And when marketers spend blindly, they don’t know their most effective channels, and they don’t know where to reinvest their money and energy in the future.  When measuring the success of SEO, look at your success year over year to see if you’re truly getting value from it.

Todd’s turn at the podium has come and he’s promised to have fewer slides than Todd while talking about SEO & SEM Combined Reporting. He is going to talk about how the two reports can be done in order to show that SEO is beneficial.

Speaking from experience, Todd says that some of the benefits of a combined strategy for your clients can include:

• Visibility, in both paid and natural results with decreases in the frequency that consumers select a competitor
• Greater opportunity to increase overall revenue
• Leveraging both paid and natural search can result in a significant lift in brand visibility
  • Up to 32% more clicks for each listing than appearing only in one or the other
• Natural campaign efforts can be prioritized based on data extracted from paid campaign
• Ability to reach entire search community

When you combined SEM & SEO initiatives it holistically yields greater results than using any one tactic along or managing them in silos. When analyzing your “SEO & SEM Battlefield”, you want to view the coverage of competitors in both the top paid and natural search positions. See what they are doing and then plan a better strategy to win.

Winning best in class means you’ll:

• Maximize ROI
  • Adjust bid strategy based on SEO ranking data
• Keyword gap analysis
  • Identify strong performing keywords missing from campaigns
  • Compare top performing keywords in paid, and try them in SEO
• Increase efficiency
  • Share landing page and ad creative across disciplines
  • Good ad copy in SEM? Try using as it as your snippet on SEO pages
• Share creative
  • Share top performing ad creative with SEO for CTR optimization
• Clickshare/SERP domination
  • Rolled up reporting identifying true combined clickshare numbers

Framework testing to maximize ROI:

• Identify top ranking organic keywords
• Select keywords for test and control groups
• Define hypothesis, success metrics and duration
• Apply testing methods
• Evaluate test results
• Rinse…repeat

In a test, they took a group of keywords that ranked in the top 3. Turned off SEM for a week.  Gathered numbers. Then turned them on for a week and compared numbers.  This is the type of test you can try yourself. Look at what numbers moved to get an idea where you need to focus energy.

Analyzing the metrics: bring in data from all available sources and analyze numbers to see where things can be adjusted in SEM or SEO to make a difference in position and spend.

Take aways: SEO and SEM do not exist in silos. These 2 things need to be compared and they need to work each off of each other in order to maximize performance.

Catfish will now take the mic for his presentation. He is going to talk about key metrics and SEO intelligence. He’s promising lots of good stuff… His presentation is titled: Key metrics (are you winning or losing?)

There are three phases of SEO data analysis:

• Performance reporting
• Opportunity identification
• Prioritization

In Performance Reporting you look at things like search rankings. Remember that search ranking vary based on localization and personalization so they are near impossible to track accurately. Also, rankings can fluctuate daily. The importance of their measure is only as important as a leading indicator that you are doing things right or wrong – or are on the right track.  Overall, search traffic and conversions are the better method of measuring success.

Use Google Webmaster Tools to get a better idea of overall ranking. It’s important to keep track, but rankings are not really the best measurement. Traffic and conversions are a much more important metric.

When doing keyword campaigns, keep these things in mind:

• Segmenting keywords into campaigns allows you to better understand performance, especially for large data sets
• Campaigns should map to business goals
• Groups map to “fat head” keywords meaning that at group level, you can see the effect of one word on other words
• Grouping is arbitrary

What is the different type of keywords? Brand vs. non-branded. What is branded search traffic? It’s any search traffic related directly to a keyword phrase that implies the user was searching specifically for your brand. Non-brand would obviously be anything else.  You should understand the ratio of branded traffic to non-branded traffic.  Regularly run rankings on all brand related phrases so that you can spot any potential opportunities or problems. Report SEO results based on gains or losses of non-branded traffic as a primary metric of success.

To calculate brand vs non-brand traffic you’ll want to grab the number of keyword phrase referrals and the number of visits per keyword. Keep track in an Excel spreadsheet.  You can then sort and total the keywords based on whether or not they are branded.

When looking at the performance of long-tail keywords, it’s always smart to look at the year over year trends in traffic in order to measure success.  For both brand and non-brand, you will have a variety of long-tail keywords to track and measure.

Catfish recommends breaking keywords into campaigns, then groups and gather a baseline for brand vs. non-brand to use as a standard for whether or not you’re growing your campaign.

Now onto the topic his been dying to talk about – SEO Intelligence and Adobe’s new tool. Adobe is offing some analytic type data that is now more affordable for many companies. SEO intelligence is a service that leverages Adobe insight to look at SEO data. Adobe insight tracks at the visitor level and there is no data roll off. The speed and accuracy of data is unparalleled and you don’t have to segment the data in advance.  Well, isn’t that just so handy.

In Adobe you can start to see attribution from branded vs. non-branded. It allows you to follow the visitor more accurately, and even gives you some off-line tracking.

The advanced capabilities of SEO Intelligence include things like cross channel attribution for online and offline.  Personalization is also tracked using triggers and company-wide targeting. There is also the ability for predictive modeling which allows you to be more proactive to getting business by knowing who is using and buying products and predicting where to market, and who to market to.

That’s it for the presentation today besides some Q & A.  Hope you were able to gain something useful from the presenters.

• Have you had enough analytics yet? Of course you haven’t. There’s never enough analytics! Here’s some just for the PPC folks.

The panel:

Moderator: Matt Van Wagner, President, Find Me Faster

Q&A Moderator: Joseph Kerschbaum, Client Services Director, Clix Marketing

Speakers:

Alex Cohen, Senior Marketing Manager, ClickEquations @digitalalex
Adam Goldberg, Chief Innovation Officer, ClearSaleing @AdamSGoldberg
Frank Kochenash, VP Client Services, Mercent
Wister Walcott, co-founder and EVP of products, Marin Software @t_wister

I’ll be honest and say that I’m probably going to understand this one even less than I usually would. I like feeling challenged by sessions but PPC is another language entirely and on top of that I’m still sick. So if this reads like gibberish, I’m very sorry.

Matt says that we’re not going to dive into the nitty gritty today. Which I appreciate.

Wister is up first. He’s very very tall.

Paid search forecasting - how they do it

• predict clicks, conversions & profit for CPA/margin targets
• bid to an overall spendting target (beug based bidding)
• find optimal CPA to balance conversion volume cost
• maximize profit
• target specific positions
• maximize revenue for a CPA target

How do you approach this?

I’m going to just leave that slide here because I don’t know what it all means.

How many keywords per group

Test for CTR and QS differences by number of keywords per group. They discovered that high volume keywords should probably get their own group.

For a CTR of 5%, no group had a QS of less than 4. Conversely for .5% CTR, no keyword had a QS above 3.

1. Above-average keyword count

2. Below average CTR/QS

3. Start with high click-volume groups first.

other useful analyses

Best practices: Any broad match terms getting more traffic than their exact and phrase siblings? Any groups with just one active ad? Are you identifying loser ads on CTR or conversions? Any budget-constrained campaigns? Adjusting bids for Day of Week and Time of Day? (Requires the time of click) Are you deleting keywords with no impressions for 13 months?

Next up is Alex Cohen.

His friend Adrienne lost a bunch of weight. How? Because the system took all the information and turned it into points that made the excess of data much clearer.

An excess of data leads to a shortage of optimization.

Slim down your CPCs with compound metrics

Look at impression, clicks, CTW, Avg CPC, conversions, conversion rate, CPA, Revenue and Net Profit.

Normalize your date by looking at Profit Per Impression (Profit/Impressions) = PPI

clck.it/ppi-links leads to great articles on this topic.

clck.it/halvarianbidding – great video on incremental cost per click.

There are 2 Fundamental drivers of prioritization – fear of loss or hope of gain.

Example: Search Query Mining – If a query is a question, the ad is the answer. PPCers job is to align those questions and answers. Not every alignment has the same value.

QueryMiner Ad Relevance Tool – useful for figuring out where you should be focusing your time.

Impression Share Run Rate: Throw money at your IS problem.

Impression share is a campaign only metric

clck.it/43tools – PPC tools galore

Next up is a battle! Adam and Frank are going to fight about attribution! Adam goes first.

There are three types of actors – introducers, influencers and closers. There can be many influencers but only one introducer and closer per conversion cycle.

All you need to do is be more accurate than you were yesterday.

The least accurate is “last click wins”.  Next is the “even attribution” – everyone gets the same credit.  Most vendors start with the even + exclusion attribution.  (There are three more “path position”, “attribution pattern” and “algorithmic modeling” but they’re not out of the box and he’s not covering them in this presentation. Algorithmic involves Ph.D.s)

Even with Exclusions

• Ability to set maximum days in path
• Ability to aggregate ad network impressions
• Ability to set hours between views
• Ability to apply “exclusions”
  • by ad type
  • by ad position
  • by time factors

They tell people that if a conversion comes from a branded search, exclude that and apply to previous steps because that’s just a re-finding search.

You may want to time out impressions — does someone really remember an ad from 30 days ago?

PPC often gets slighted by the last click model.

Use exclusions to zero in on the terms that are actually bringing you traffic so you can ignore the navigational searches and bring traffic to the terms that actually drive leads.

Frank is up last.

He’s going with a nuanced metaphor about snake oil. Okay then. Snake oil originally was Chinese medicine and it’s still used. The problem is not the oil, it’s the misinformation that surrounds it.

Today’s takeaway: Measure the algorithm.

Advanced ROI attribution is about predicting probability of conversion.  How is it used? Providing weights used in calculation of ROI.

Viewing the customer journey as an evolution of conversion

In a perfect world it looks like this:

Realistically it looks like this:

Some pitfalls:

• assume the influencers are static
• assume every exposure or user interaction deserves credit
• assume every cookie/consumer is the same
• don’t adequately handle external influences
• assume creative and placement don’t matter
• assume…something else. And more as well.

You need to measure the difference between predicted and measured to find the model error (and the model error of time. You want your model error to get less over time and you want it to be adaptive.)

Buying ROI attritibution algos

1. Request model error rates prior to launch and over time
2. Ask how does the algo adapt to disturbances or changes
3. Ask how it adapts.

How do you use product-keyword algorithms for a particular targeted case?

Take ads out of circulation when a product is out of stock, put them back in when it comes back into stock. Use it to drive ads by SKU. Look at how correlated the SKU is to the keyword. (There may be variations like sizes and colors.)

[There's a chunk missing here while I swapped my battery. My apologies.]

This is not an 80/20 situation: most of the value requires most of the work.

When buying product-keyword attributions algorithms:

1. Request correlation statistics (i.e., how do you know it’s working?)
2. Ask how do you establish correlations between things I can control (like keywords and LP) and things I can measure (like product inventory)?
3. Ask what data is being collected to enable these calculations?

Above all, measure the algorithm.

The Q&A is most very specific so I’m skipping it.

Good morning, my friends. Welcome to the first morning of this super quick but super awesome two day conference that we call SMX Advanced. It’s the fifth anniversary of this conference and my, how it has grown.

It’s gray and just a little bit rainy here in Seattle but inside it’s all good times and sunshine. That’s because we have a fantastic panel about to discuss my eternal nemesis — Analytics.

Oh, analytics sessions. Why can’t I quit you? I never understand but I always attend.

Here’s the line up.

Moderator: Chris Sherman, Executive Editor, Search Engine Land

Q&A Moderator: Robert Cooley, CTO, OptiMine Software

Speakers:

Madan Bharadwaj, Vice President, Visual IQ @madanbwj
Allison Hartsoe, VP Analytics, Semphonic @ahartsoe
Phil Mui, Group Product Manager, Google @philmui
Khalid Saleh, President, Invesp @khalidh

Madan is up first. He makes sure we’ve all had coffee first. Good man.

The science of attribution is about looking at the entire consumer experience.

Today’s themes:

• Clicks as a marker of “intent” during the experience
• Time Lag from various point in the experience.

These have in common the need to look at the entire experience, not just the last click.

Entire consumer experience = entire engagement stack.  They look at every single user, not just converting users.

“A click is not a click is not a click.” — A click on display is not the same thing as a click on search. In display, it makes sense to measure conversion rate against clicks because you’re paying for it. Clicking on a search query is very different. You need to analyze the holistic experience to understand what a click means in each channel.

Search is a pull medium, not a push medium. It’s a very different kind of action and intent.

The online population is a sample of the entire population. It’s reasonable to assume that when it’s time for them to express their intent, they will actually come looking for your brand.

Time Lag: The traditional time lag definition is flawed. It’s not just the time from the first search to the final conversion. There are several channels.

Some conversions are clicks only (from search), some are impressions only (display), some have a mix. Looking at the engagement stacks, they found 5 path markers.

What does each time lag tell you? People with just one click converted fast, within an hour. People with more than one click converted on a longer scale. Very different behavior and based on their expectations. If you know that people who see an ad convert in a different browser session more than a day later, what would you tell them in your ad?

For this client, they found two groups — instant gratification group and delayed gratification group.

The key is to analyze and optimize the whole experience, not just the last touch.

Khalid is next. His company makes conversion optimization software.

In order to increase your conversion rates, you need to know where to start the optimization.

Plan where to start, Implement, Iterate

So where should you start?

There are more than 12 different factors.

The basic choice is  the home page, the high exit rate pages, the high bounce rate pages. Usually the shopping cart is a suspect.

Look for trends, not numbers in analytics.

Look at the number of visitors that land on your category pages but never make it to the product pages. Get the pre-product page abandonment rate to find out how many people are never even getting to the point of adding to the cart. (Having an add to cart on a category page is almost useless.) Try for a PPAR of 25% or less.

A category page exists to funnel traffic to product page. It helps customers select the right product for their needs (product, navigation, comparison, feature selection, etc.) Categorization is complex and usually set at the beginning and is often hard to change.

One of the problems is if they are jumping back up categories instead of down further into product pages.

Keep an eye on the bounce around affect. — the number of visitors who navigate to a parent or sibling category instead of to a product page. Keep it less than 20%.

Product page effectiveness rate: # of visitors who click on the add to cart button/total visitors who visit the product page

PPER Goal: 10-15% (higher for lower price points, lower for higher price points.)

Wow, he went fast.

Allison steps up next. She’s certified Google and Omniture (Abode).

What are the hidden dimensions of Google Analytics? — use the ROI Revolution GARE greasemonkey script on Firefox to get lots more information and options.

V5 of Google Analytics has some of these dimensions but not all of them. The script doesn’t work in V5.

How do you track social media traffic?

The quick and dirty way:

• Can filter incoming source to report Twitter or Facebook.
• Create a custom segment to apply across reports

The better way:

Use GraphEdge for Twitter for Follows, Unfollows, see impact over time. It’s an external tool. Starts at $5 a month.

Best Practice: Tag your links w/ campaign codes and use URL shorteners. That way you can track conversions, last touch, 6 month cookie.

Tracking within Facebook:

• Facebook Google Analytics Tracker (FBGAT) extension
• Free, image tracker per page (contact, product)
• Track fan vs non-fan, funnel, create goals.

How to get more out of SiteCatalyst?

Must declare a campaign identifier in admin settings, (e.g., “cid”) and tag campaigns with it. Make sure you add it to every single new campaign.

Filter by marketing channels. It’s a post-processing playground. Changing things around won’t hurt anything and it’s flexible (but not as flexible as Google Analytics)

Phil will be the last presenter.

Two feature announcements in this presentation.

Four categories in GA – Web, Rich Media, Mobile, Social

To get the most out of GA, work with a certified partner. [AD!]

Topics:

• Mobile
• Social
• Funnel
• Search

Mobile SDK – easy to set up, two lines of code. Tracked data is queued and dispatched in bundles.

Tracking user preferences is also two lines of code. [I'm sorry this is very sparse, he has a lot of slides and I can't capture them in time.]

Tracking in app purchases – e-commerce tracking

Social

Feature Announcement – GA V5 is an evolving platform, the newest is tracking for social buttons on site. Every social network and button type will be supported with modest setup and +1 will be supported natively. Visit bit.ly/gasocialplugin to sign up.

The last click just isn’t enough, you need the whole story.

How much should I be paying my affiliates? — Assign accurate value to CPA

Feature Announcement two: Google Analytics and GWT will be integrated starting this month so you can get GA information in GWT. [The room cheers.]

In the comments, Asaph notes: Minor correction: starting this month you will be able to see GWT information in GA, not “GA information in GWT” as was written here.

Mea culpa.

Not new news but still interesting, there have been some screenshots circulating around some of our favourite SEO blogs of Google testings some new interface designs. This interface testing seemed to happen around the start of the month and there were a few different designs spotted in the wild. One of the key features across most of the examples we saw was white spce, a lot of white space and where there are site links the gap between position 1 and 2 in the SERPs is gigantic.

We did some checking here at our office and couldn’t seem to replicate the results pictured above.

Google likes to test tweaks to their interface, it happens fairly regularly but in different places around the globe. If you like trying out new ideas Google has cooking, head over to Google Experimental and join some experiments. Don’t get too attached to a new feature you’re trialling though, it might not make the grade.

Bing Business Portal

Bing recently released the Bing Business Portal, their answer to Google Places. And it’s actually pretty cool. Some of the features you get with the service are:

• Free customisable mobile website
• Social media links
• Dedicated logo (that doesn’t get confused with other uploaded images)
• Free coupon hosting that has a chance of showing up in m.bing.com searches
• Sliders to indicate what your specialty is within your industry
• Smooth verification service (reportedly better that Google’s)
• Other features that match Google Places (Name, Category, Address, Phone, and basic contact info, QR codes)

With all that stuff for free how could you afford not to sign up?

Google Analytics adds Page Speed reports

Google has been offering free tools to help webmaster optimise page load times for a while now in the form of a Firefox add-on for Firebug called Page Speed. They also released a version for Chrome and then an online version.

The problem with all those options is that they are limited to checking one page at a time. So if you have a website with a few thousand pages, chances you’re going to spend the time to check all those pages individually for page load speed stats is pretty slim. Enter the Site Speed report in Google Analytics. It can help you identify which of your pages are the slowest, which traffic sources lead to faster page loads, which browsers your pages perform better on and how geography effects pages load times.

How do you make it work? All you need to do is add a line of code to the Google Analytics JavaScript, the details are here. Also, to see Site Speed stats, you’ll have to use the new version of Google Analytics by clicking the link in the top right then look for Site Speed under the Content menu. If you don’t have the code installed, the charts will be blank.

And that’s all she wrote for this Hot Tub.

Yes they can coexist, that is SEO and Usability and very often in a complementary manner; the data sources are just different. We encourage usability folk in all our clients to get more involved in understanding not only SEO, but also what Analytics is telling them, along with their focus groups and other user data.

It is often a case of knowing what you know for example. You know traffic went up due to an event, maybe an ad campaign running at the time and that the increased traffic was most likely on branded terms. Secondly, you know what you don’t know which is exactly what keywords drove traffic last week, but you know where to look, in Analytics. Thirdly, there is often the part where you don’t always know what you don’t know and unless someone tells you or points out the obvious, which may not always be that obvious you will never know. This data is usually available from a number of sources; you just need to know where to look and what you are looking for. Things like which pages IT removed and didn’t redirect, what content is converting new users the most and so on.

Combining Analytics with SEO tactics and website usability can be highly profitable if you do it right. The biggest challenge is knowing what to look for with respect to your business objectives, as well as getting the various parts of larger organisations involved and coordinated around this.

You need data

Assuming you already have Analytics set up, and some sort of ranking monitor, much like our very own SEOToolset, you already have a lot of valuable information at hand.

But first things first, how much do you know about your visitors?

• Do you know where they come from?
• Do you know what platform or device they use?
• Do you know where they live?
• Do you know when they visit your site?
• Do you know why they visit your site?
• Do you know which parts of your site they like the most?
• Do you know which keywords are used to find your site?
• Do you know which keywords are most relevant to your users or customers?

Do you know which of the above are interesting and which are highly relevant to increasing profitability? All of the above you should be getting from your analytics software, and are physical processes, easily measured. In addition you can also start to measure the user journey to a limited extent, prior to website entry, i.e. which keywords were searched on, what social network referred them etc.

Understanding the demographic of your users i.e. male or female? Income etc. can be harder. Even harder is getting a handle on their decision making processes or cognitive processes. Everyone is different and conversion a site occurs at an individual user level. Have you ever seen a committee or a group surfing a website? Measurement of cognitive processes is very difficult and time consuming.

Developing Personas

When working on a site, understanding your target market is essential. Who are the majority of your users? Appealing to parents with toddlers is very different to appealing to middle aged men and women running a small business. Also the value and nature of your product or service will further impact on personas, and how they conduct their activities on your site.

Personas are important, as you will need to consider user paths within your site and how you address these personas. Broadly, 4 personas exist which need to be taken into account. These include:

1. Spontaneous
2. Competitive
3. Methodical
4. Humanistic

You can listen to a podcast by Bryan Eisenberg here on personas.

Approach

While SEO must have its own methodology and approach, likewise usability and reporting, getting the most out of your Analytics and other available data will require an integrated approach from different disciplines and a process. As a result, it is very often an activity that can slip through the cracks.

First of all you need to define success. What are the outcomes or the ROI this process will need to drive, and who will own it. Your SEO team is often a good place to start, and revenue earning activities will get attention and budget.

Developing the opportunity by understanding your audience and data mining is key. Getting the data is very often the easy part, interpreting the data and making recommendations supported by data is where many organisations need some help. Testing is then critical. Not all hypotheses and related tests will work, some might even fail but that is why you test. The tests that are successful you may adopt and all of this will need to be managed, from design through to implementation.

Ensuring the process is properly managed with a change register, listing the outcomes of testing and then a Steering Committee or some other formal process to sign off on which changes become permanent fixtures

5 Things to Watch

At Bruce Clay in Australia we review out client’s analytics accounts as part of the “Optimise” phase anytime from weekly through to quarterly, depending on the type of site and the definition of conversion. Part of the checklist we use when reviewing client data and where we find quick wins to improve usability or conversion include:

1. Underperforming content

Use the Traffic Sources, Keywords submenu item and compare the current period to a reasonable prior period. This period could be a week, a month a quarter, depending and we say reasonable as obviously if there have been public holidays or other events the data may be skewed. Look at keywords where there has been a substantial % change with a reasonable number of visits. Investigate any large change changes to establish the reason and any related opportunities.

Check out the pageviews and bounce rate on these keywords. Mismatched content often has low number of pageviews or lower than average pageviews and/ or a high bounce rate and can often be tweaked on the page or via a new page feeding into your content strategy to leverage these opportunities. Identify underperforming content and set goals around this content and how it may improve conversion.

2. Underperforming keywords

If the 80:20 rule applies to the keywords driving traffic to your site, take these as well as any with significant changes identified above and download the Adwords keyword volume data to compare the traffic you are getting versus the total monthly local search volumes available. Establish your market share, look at your existing rankings to see if this is reasonable and identify new opportunities where there is clearly further market share available with respect to traffic.

3. Reclaim lost links

Log in to your Google Webmaster Tools account and review the crawl errors. You can find these in the Diagnostics section under Crawl Errors. Check for 404 pages in the linked to pages. This in certain sites may be a lot of work due to volumes, but there are very often gems in this list, where a promotion or page has been removed and not redirected losing valuable PageRank. Get these crawl errors fixed.

4. User paths and conversion funnels

Map out your users paths for the various personas or an average user for your site. This may be a consolidated view of a very singular path, but ultimately should lead to an action. Once completed, using analytics see if there is any correlation between how users broadly interact with your site as opposed to how you think they should use the site. Try to establish where they go astray in the process. You cannot plug all the holes but think of it as gently herding cats through your site.

Then, if you have a conversion process, work carefully through each step looking at where users go, the time they spend, how many are lost etc. and at what stage. In Page Analytics can be useful but is limited. Heat mapping software is a far better option to see exactly what users are doing.

5. Underperforming conversion

If you have ecommerce tracking installed review each section and check for products, contents or keywords which are converting significantly lower than others. Establish the reasons why and if any further opportunity exists.

Conclusion

Analytics is not something that just tells you how much traffic you are getting. How you use it however and the value you gain from these insights is up to you. Access to Analytics for the masses such as Google Analytics is somewhat recent and many users are still trying to figure out exactly how to use the information gained. The sooner you start incorporating Analytics into your SEO and Usability processes the quicker you will start getting results.

Analytics sessions, like legal sessions and eye-tracking sessions, are my catnip. I just can’t resist them, no matter how much I try. Which is why we’re here in our second analytics session in a row. Ready?

• Moderator:
  Nancy Keene, Associate Media Director, Digital MV42° | MediaVest
• Speakers:
  Fernando Rodriguez, CEO, Terra Networks
  Paul Lima, CEO, Lima Consulting
  Armando Rodriguez, General Manager, Yahoo! Hispanic Americas

Now that you’ve got the players, let’s get right to the game. (Opening Day at Dodger Stadium is 8 days. Oh yeah.)

Paul Lima is up first.

Digital Marketing is the new finance

Why Analytics?

Customer Insights and actionable intelligence

Good job on coming into this room instead of the Facebook room. They’re still trying to figure out how to make money, we’re actually learning how to do it.

We’re moving beyond Caveman Analytics (see last session)

Popular metrics are not always the most effective. Don’t do time. Analyzing data in aggregate is a crime. Segment then analyze.

“Begin with the end in mind” You need to know what you’re measuring before you start.

5 tips on segmentation

Needs Based Segmentation

Language is easy and obvious. Need — security, power, esteem, change — is harder and less obvious to target.

Use segmentation maps

Use psychographic segmentation for professional services

– Validators are the most valuable statistically.

Align your segments with your product portfolio

Use psychographic profiles for online purchasing segments

Case study: school for struggling teens

Who are your customers? They thought they were pitching to the moms – soft, embarassed, worried. But the decision makers were the fathers – professional, direct, aggressive.

Barbara Billingsley versus Gordon Gekko

Hispanics Engage even more than non-hispanics

[and he just blew through about five slides -- yikes]

Fernando Rodriguez is up.  He makes a joke about the snow… WHAT SNOW?

What’s the difference between the general US market, the US Hispanic market and the Latin American market. LatAm is slightly behind in the use and adoption of analytics tools

Forward looking view:

• Mobiles and connected devices: can we keep up?
• Do we understand the new landscape of mobile operating systems and aqpps?
• De-emphasize click and web measurement/should focus more on the consumer behind the mouse
• The privacy debate versus the benefits of a personal experience.

-US Hispanics spend more $ per ecommerce transaction vs. non-Hispanics. More receptive to online advertising & new digital media experiences.

- LatAm Internet Users are 24% more likely to use social networking vs. avg. global internet user

He swiped his next slide from a MorganStanley presentation. Hee.  Basically there’s a huge global opportunity that we’re still missing out on. US Hispanic markets could offer 1B in revenue.

Armando Rodriguez (who recently joined us on the blog) is up next.

113 dollars/user vs 6-7 dollars/user for us hispanics.

TV is king in the Latin markets, even in the US.

How do we shift those dollars from offline to online?

How do we bring together the publisher and the advertisers?

Science, Art and Scale

Science gives us the data that we need but how do we leverage that to provide innovative and personal experiences? Can we scale it?

Create deeply personal experiences

• 45,000 different versions of the Yahoo! home page based on the user’s need and interests.

Drive Scale with Insights

• Allow for navigation in Spanish and English as well as advertising in both.

Use facts to create understanding

Pre-Sale knowledge drives analytics during the campaign drives recommendations for solutions drives pre-sale planning in the future

Key Takeaways: Yahoo’s POV

• Insight, not Hindsight – use facts to create understanding Latino audience
• Unlock creativity: interactivity – build lasting engagement
• Touch meaningful numbers – reach right audience with right message

There were cream puffs for dessert today. You know what there wasn’t? Bagels.

Yes, this is a thing now.

But onwards and upwards, ladies and gents. It’s time for more analytics. In case you didn’t get enough Matthew Bailey yesterday (and really, who could get enough Matt Bailey?), he’s here with moderator Mark Jackson, SEW Expert & President/CEO, VIZION Interactive and fellow speakers Ulli Muenker, Search Marketing Manager, Bloomberg / BusinessWeek and Jim Yu, Founder & CEO, BrightEdge. We’re talking SEO metrics that don’t suck.

This session is packed. Because we all know that we’re terrible, horrible people when it comes to tracking what we do.

Ulli Muenker is starting us off. I’m excited because she’s the first female panelist I’ve seen so far at this conference. She’ll be presenting the redesign of Bloomberg.com as her case study today.

Challenge: Opportunity to technically optimize website for SEO but C-levels want to know the $ value before they invest resources to it.

Factors that skew results:

• Time -when will the changes kick in?
• Content – Are content changes/updates happening at the same time going to affect your optimization?

C-level doesn’t want to hear “we don’t know”.

Bloomberg’s Redesign:

Time: All technical changes launched at the same time

Content: No changes made

Basically a test environment and done all in a very tight 6 month timeline from 12/09 to 6/10

Because of the timeline, when they wanted SEO in the project plan, they just got a blank “…uh, maybe later or as a separate project?” [Room laughs]

What do you do? Get in good with the tech team and talk them into ‘since you’re doing this anyway’. Make friends with them while they’re working so that you can incorporate SEO that way even though there’s no official time for SEO.

1. Customized meta data for all nav pages – Worked with an SEO copywriter for this. Had an expert work on it rather than wasting time herself 
   
2. Redirected all duplicate URLs to on unique version – sometimes repeated 10 or 15 times because of JS issues and other reasons (parameter was added based on the position on the page for example). Changed to just one URL.
3. Eliminated duplicate URLs & keyword-rich URLs
4. Added KW-rich header tags – there were no heading tags used at all

You can see a clear lift on the implementation of SEO and more on the launch of the new site.

CEOs want real numbers however, so figure out your traffic, revenue and the SEO share of it. Show the changes over time month over month to prove value. Compare your before and after.

Summary of the impact of SEO: $1M/year extra just with minimal technical changes on the site — this proves SEO value.

Results: Traffic Increase, Sustained Traffic, More Buy In internally

[The wifi is feeling the strain.]

Jim Yu is up next.

Things to know first: SEO is different than other channels

• Touches many people across the company
• Audiences are varied and has different needs – from C-level to directors to technical to PR
• Varying levels of SEO sophistication requires metrics easily understood throughout the company.

5 Ways to get Meaningful SEO Metrics

Tip 1: Identify One Key Business Metric

One metric simplifies communciations. Focus on a key driver of value aligned throughout the organizational goals.  Ex: Purchases, Applications, page view (for media sites), sign-ups. Pick something that makes you money.

Tip 2: Separate out Brand and Non-Brand

They have very different SEO behaviors. Brand searches are very tied into other channels in the company.

Make sure that you don’t have any leaky conversion funnels.

Tip 3: Use share of voice to compare to competition.

Figure out who in your niche has what percentage of marketshare. Share of Voice = SUM for keywords (search volume x click-through)

Tip 4: Have a Phased Approach to Share of Voice

1. Count listings – 1 listing, 1 point
2. Use ranking as a scoring function – more precision
3. Combine with search volume to estimate visits (Google Keyword Tool)
4. Combine with conversion to estimate value – ideally from your analytics data

Tip 5: Star with high impact intiatives

Examples:

• Optimize landing pages
• Find high value keywords on page 2
• Target keywords from upcoming PRs
• Fix structure issues

And now Matt Bailey.

He’s a big fan of making things more complicated. Edward Tufte said “To make charts simpler, add more information.” Matt adds “But, don’t give that chart to the CEO.” Hee. Truth, Matt.

More information gives you more insights.

6 Keys to Integrating

Show Multi-variate Data

1. Show comparisons

Show Casuality — Ask Why?

Show Different Reports for Different Sorts — give them the report that they need. Give the CEO a big colorful report that he can color in.

Show the Money

Get beyond Caveman SEO (it’s related to hamster analytics. hee.)

More data points = more comparison

What’s your first goal? To make more money. We’re not here for a noble goal like saving pandas. We’re all here to make more money.

People don’t get deep enough into Google Analytics. Set goals. You have to set goals.

Goal set, then by source. (Pivot tables are beautiful! If they’re not to you, add more data!) Then look at keyword and landing page. Now you’re not looking at ranking, you’re looking at money. That’s much more important than ranking.

Once you have all of that, you get goal conversion rate and per visit goal value.

Better comparisons lead you to better casaulity. You can segment your acquisition channels and see what’s providing you the best value.

Look for quality over quantity. Who sends you the best visitors. Who converts, not just who sends you the most traffic. Look at where your links are coming from and figure out what makes them qualified. Include a screenshot of the source link so you know what they’re seeing.

Twitter provides links that provide zero context so it doesn’t tend to have good conversion. You don’t find out what you’re getting until you get there. Discussion forums provide highly engaged and highly informed visitors. It’s one of the most overlooked social sources. Target the most effective things first.

Look at the page at the same time as analytics. You’ll find things that are broken and why.

[I highly recommend combining this presentation with Matt's presentation yesterday. If you haven't read that one, you're missing out. Literally, there's a slide here I'm not recapping because he explained it yesterday.]

Tracking Analytics + CRM

Learn that high quantity leads do not equal quality leads. Learn that high quality leads die at idiot salesmen. Figure out what information the salespeople need to have on hand to close the deal and get it to them.

Outside factors

• customer CRM
• Email promotion
• brand promotions
• catalogues
• content feeds
• customer service