Performance Marketing Tips

Let's Get Down to Business

Posts Tagged ‘optimization

The Open Campaign, Transparent Data Makes for Awesome Marketing Idea!

leave a comment »

the open campaignI was reading Billy’s Blog earlier in the week and he had an interesting post about something called The Open Campaign.  It’s a joint promotion between Widemile, WebTrends and a few others (not totally out of left field WebTrends just purchased Widemile).  In any case the idea is for these conglomeration of online marketing and analytics experts to post their campaign methodology and results publicly – very cool!!!

Billy talks about some of his initial testing ideas with Widemile – nothing earth shattering yet but practical application is nice to see.  They are also using ForeSee as their survey vendor which I think is a great idea.  I used to be extremely skeptical about the use of opt-in surveys in conjunction with analytics data but I’ve come full circle in the last year.  We layer in survey results with our click tracking behavioral analysis to really create a powerful optimization roadmap.  I’m looking forward to seeing how The Open Campaign uses the survey data.

Be sure to check The Open Campaign often for updates, I’m looking forward to learning a lot from this project.  (and BTW, what a great marketing idea!)

If you’d like to learn more about optimization pick of a copy of Landing Page Optimization For Dummies (For Dummies)

landing page optimization for dummies


Written by Greg M

July 31, 2009 at 6:17 pm

Optimization KPIs for Testing a Checkout Process

leave a comment »

funnelMulti-step checkout processes and conversion funnels present some unique testing and optimization challenges. Most complex conversion processes like these track campaign and website success across multiple dimensions – conversion rate, number of items bought, total dollar value, gross profit, etc. In a testing environment those items are sometimes way downstream (I once worked on a site that had 29 steps – yikes!!!). Using both a primary and secondary KPI can help you achieve statistical significance faster in your testing and optimization program.

A continual challenge for site side multivariate and A/B testing is that statistical significance can be tough to reach in a reasonable length of time – 2 to 4 weeks is the general range. Using a primary KPI like click through rate on the test page will yield a much greater sample size since it’s further up the funnel. In most cases it’s substantially faster than waiting for an adequate sample size to shake out downstream.

The secondary KPI will be used to validate that we’re sending quality traffic through the funnel with our optimization program and not just tricking people to click on a bright shinny object. In this case we could use total dollar value and compare that against our website’s historical benchmarks. Statistical accuracy isn’t as critical at this point, we just need enough conversions to see if our optimized page is in the ballpark of the norm – 80% confidence or less would be fine if you see enough life on your primary KPI. If these validation KPIs are way off then we need to let the test run longer to gain more significance and start analyzing the pages to determine why we’re generating a lot of low quality clicks. Overall this should greatly boost the testing cycle time of your optimization program.

To learn more aboiut web analytics and KPIs pick up a copy of Advanced Web Metrics with Google Analytics from today!

Written by Greg M

July 24, 2009 at 10:59 am

Posted in Tips

Tagged with , ,

Book Review: Landing Page Optimization: The Definitive Guide to Testing and Tuning for Conversions

leave a comment »

landing page optimization handbookLanding Page Optimization: The Definitive Guide to Testing and Tuning for Conversions

This book is a big win for Tim Ash and all interested in landing page optimization and testing as I believe it’s currently the most comprehensive and well thought out book on the subject. Not only does Landing Page Optimization cover all of the necessary tactics for creating, launching and interpreting A/B and multivariate tests but Ash thoughtfully examines two critical aspects of human relationships and their impact on your testing program – first, the different personality traits of your site visitors and how they change based on their current role and second, how to work with all of the critical roles (IT, QA, brand managers, creative teams, etc.) whether your optimization program is for internal or external clients.

1)  Here are 5 key lessons learned, but remember to pick up your copy of Landing Page Optimization so you don’t’ miss anything.

2)  Everyone may want equal space on your website but they don’t deserve it. Your landing page should be modified to best serve the mission critical visitor class

3)  If your landing page is designed for a single conversion event the main navigation or structure of the corporate site should be removed.

4)  ADIAS – awareness, desire, interest, action, satisfaction – know the stage of the buying process for particular pages

  • In the Awareness stage all key information needs to be above the fold
  • In the Desire stage more detailed information can be supplied and visitors are more likely to scroll below the fold

5)  If you’re focused on testing elements of your offer (price, promotion, etc.) then traffic sources are important. If the focus is more on functional usability issues then traffic sources are less important. First optimize functional issues across all site traffic and then fine tune elements for different traffic sources.

Always get a project-based support commitment from IT and have a dedicated IT resource throughout the program.

Pick up a copy of Landing Page Optimization: The Definitive Guide to Testing and Tuning for Conversions
on today!

Written by Greg M

July 22, 2009 at 12:22 am