Analytics Action Plans For PPC & SEO
Moderator: Christine Churchill, President, KeyRelevance.com
My last lunch of SMX West hit the spot. This session is sure to do the same. Dennis is up first.
He wants to challenge us to think of the goals of your site before you set up an AdWords strategy. They categorize customer needs with some of the first questions they ask them. How many hits can we get? Not a great sign. How can we get more data? Also needs coaxing to get them thinking in the right direction.
What is User Engagement?
- Turning on a prospect and surrounding them with useful information.
- A deeper understanding of site visitor behavior and intent.
- Rarely possible in one metric or KPI
- Beyond: UX, conversion tracking, or time on site
- Beyond satisfaction: it’s a measure of interest and action and may be represented as loyalty
- Importantly, it requires a plan to measure effectively and affect improvement
- It’s rare to find a company satisfied with their analytics plan
- Too many confuse reporting with analytics
- Too many suffer from analysis paralysis
- You don’t know what you don’t know, and that’s the biggest challenge. Figuring out your marketing goals 6 months or a year from now is hard. But you don’t want to lock yourself into a solution that can’t change with your changing needs.
Analytics Planning Strategies
- Do you have access to good data
- “and” strategy, not “or” (Reflective of Avinash’s comment at this morning’s keynote)
- There is no perfect or complete tool
- Conversion metrics (Think of soft conversions as well)
- Reports are not analytics
- Ad Hoc Exploration
- Plan to fumble around. That’s how you learn.
Here are some search KPI examples (note: all KPIs assume across time)
- SERP rankings for targeted keywords/phrases
- Competitive organic search share for top 200 industry search terms
- Organic search traffic trending
The info in the slide above doesn’t show you compared to what, over time. Plus with new technologies, a single KPI could change meaning over time.
- User engagement KPIs
- Average PVs per visitor from search
- Conversion rate from paid search
- Bounce rate against “competitive benchmark”
- Review comp sites
- Downloads from socially referred visitors
- Encourage sharing
- Form completions (leads) from organic
- Enhance conversion opps on top organic pages
Instead of looking at ranking, it can be better to look at “Breadth and Depth”:
- Breadth=total traffic from organic search
- Depth=total number of keywords from organic search
Ryan is next. He says there are lots of metrics that can be measured and he just threw out 20 different acronyms. So much data, so little time. So what to do about it?
- Stare at your screen
- Take a stab in the dark
- Or take a step back and audit the situation.
Did you say audit? They take a long time and can be painful. But it gets results. So what to expect?
- CPC -74 percent
- PPC CR +264 percent
- SEO CR +250 percent
- Leads +40 percent
- Cost per lead -40 percent
- Orders +54 percent
With an audit you come up with an actionable plan based on analysis of results.
The PPC scorecard:
- unique KPIs
Here are 20 questions that speak to 1 and 2 above.
1. Number of PPC accounts
1=1, 2=2, 3=3, 4=4, 5(+)=5
2. Number of campaigns in an account
1=1, 2=2, 3-5=3, 6-10=4, 11(+)=5
3. Number of ad groups per campaign
1=1, 2-5=2, 6-10=3, 11-19=4, 20(+)=5
4. Total number of keywords across an account
0-100=1, 101-1,000=2, 1k-5k=3, 5k-20k=4, 20k(+)=5
5. Number of active keywords per ad group
1=1, 1,000(+)=2, 51-1,000=3, 6-50=4, 2-5=5
6. Number of text ads across entire account
1=1, 2-5=2, 6-10=3, 11-19=4, 20(+)=5
7. Active text ads per ad group
1=1, 8(+)=2, 5-7=3, 2-3=4, 3-4=5
8. Unique landing pages
1=1, 2-5=2, 6-10=3, 11(+)=4, Infinity (multivariate testing)=5
9. Number of goals (conversion events)
1=1, 2=2, 3=3, 4=4, 5(+)=5
10. Bid rules
0=1, 1=2, 2=3, 3=4, 4(+)=5
Non-specified=1, account level=3, campaign level=5
12. Keyword match types
Broad=1, broad+negative=2, exact=3, phrase=4, embedded=5
13. Conversion tracking
Search Engine=1, Customized SE=2, Google Analytics=3, Customized GA=4, Paid analytics=5
14. Campaign update frequency
Annually=1, Monthly=2, Weekly=3, Daily=4, Intraday=5
15. Call tracking
None=1, One TFN=3, Multiple TFNs=5
16. Bid management
None=1, Manual=2, Conv. Optimizer=3, 3rd party=4, 3rd party w/ attribution=5
17. Landing page testing tools
None=1, Manual=2, Free (WSO)=3, Customized WSO=4, Paid w/ Segmenting=5
18. Campaign settings
None=1, Location Based=2, Network Selection=3, Bidding/Budget=4, Advanced=5
19. Text ad copy
DKI=1, Static=2, DKI+Static=3, DKI+Static+Custom Display URLs=4, High KW Dense Converting Static=5
Never heard of it=1, w/ Text Ads=3, w/ Display Ads=5
Now tally up your score.
Rich takes the podium next. They focus on the intersection of creative and data. You can be both, and here’s a new hybrid approach.
- Goals driven analytics
- Beyond the conversion
- Monetization modeling
- Some examples
When we succeed with clients, 5 characteristics pop out
- They know analytics and technologies
- They seek first to know their business and goals
- They seek to improve access and appetite for data – upward and across the organization
- They focus strictly on actionable data that empowers stakeholder decision making
- They don’t avoid the weeds, but they don’t get stuck in them either
Identifying stakeholders :
- Ask really good questions
- Identify their unique business objectives
- Roll up stakeholder feedback
- Define collective business goals
- Review and seek consensus across stakeholders
Business goals dictate site goals, which inform other digital channel goals.
- Less is actionable
- More is nice to know, or at worse, paralyzing
- Okay to have sub-goals
Looking Beyond Conversion
There’s a lot of site traffic but only a small portion of that is measured as success. All the other activity also has success, so they try to understand the value of it. They build models that define the money value of micro-conversions. They are custom-built performance models.
Use monetization to guide project or optimization priority. Score each project for size of opportunity, ROI, and business priority. Weight each score to determine an aggregate weighted score. Use score to guide decisions, not as an absolute rule. As revenue / cost estimates are revised, so is the monetization model.
What’s the opportunity cost of in-action? This is very effective for justifying longer-term, higher-cost projects.
Ian is our final speaker. (BTW, check out our awesome interview from this week’s SEM Synergy!) He says he’s going to get into the weeds. Who has seen a raw log file? A good number of people have, so he says this shouldn’t be too awful.
If you stick your head in the sand, the only thing you can out of is your bum. With organic search, worry about three things:
Opportunity and competition are about optimization, improving rankings on terms you want. Attribution is about keeping your job, showing evidence that what you’re doing is working, and if you need to do more of one thing or another.
Keyword driven research isn’t bad, but you have to look at other things first. You’ll miss opportunities for keywords that are almost on page one. You’ll miss the opportunity gap. If you can follow that up with a narrow competition gap, that’s a good position to be in.
Step 1 is to determine opportunity. Start by looking at pages, which pages are getting the most pages from organic search. For each of those pages, look at which organic keywords are driving traffic to those pages. Now look at keyword data. If you see 2,900 searches for your keyword, and 250 visitors from that term, there’s an opportunity gap. The question now is, can I compete. He recommends the Keyword Difficulty Tool or SEO for Firefox that show you how hard it is to compete on a term.
Now you can optimize the ranking page. Link to newer content to drive people to updated content on your site. A tool that will show clicks from page 2 and one that will show pages with no clicks, will help you.
The only way to do attribution is to learn to love your log files. Scrub the log file so all you have are pages, and sort traffic by cookie and IP address.
I betcha Ian will post these on his vimeo!
[Ian also shared his Google Analytics Cheatsheet. --Susan]