DISCLAIMER: This post is written as a live blog from Conversion Conference. There may be typos and grammar to make my high school English teachers weep. Please excuse those … it’s a fast-paced conference with back-to-back sessions and no time for proofing or even proper writing.
Full coverage of the event can be found here.
Now we’re chatting about Distilled with a session, “The 7 Tactics That Will Turn CRO Experts into SEO Winners of the Future”.
After telling me that SEO’s will soon be out of a job …
He talks about some test Google is running including asking people how they like results and their use of user clickthrough rate tracking. They’re testing user behavior so if you’re not you’re about to lose some rankings. They’re also tracking pojo-sticking (that is – a user bouncing between results and then selecting one and staying).
Their constant testing and tweaking of algorithms based on user behavior and the rapid nature of their data collection allows for quick reactions and changes on results and algorithms. My own note is that with AI enhancements we’re sure to see this speed up and get accentuated.
There seems to be a correlation between bounce rate and rankings. There is also a correlation to links but he predicts the death of links as a ranking factor. I disagree based on the announcement at http://searchengineland.com/now-know-googles-top-three-search-ranking-factors-245882 but the point that user experience is critical is not lost.
He points out that even if CTR isn’t a ranking factor the testing and tuning will only increase traffic based on increased traffic for current rankings.
Additionally you would also be looking at improving bounce rates and user experience. This also has benefits whether it helps rankings or not.
Essentially he accurately points out there is no losing in this testing and tuning.
Understanding The Customer
Focusing on keyword only does not necessarily hit the customer. Interviewing customers and other actions allowed them to create personas. They discovered the client was great at one persona but others were totally confused. The site didn’t answer questions so they lost them.
By creating the pages necessary to answer the questions (10 to 12 pages) and significantly improved the conversions an engagement by being able to appeal to new personas.
Sell The Value
If you’re not tracking revenue then data is not complete. And framing the data in the long term can help clients feel the need to do more of it.
Structured Approaches
Average large sites wait 6+ months to implement changes. This is leaving money on the table. That’s because the SEO process. A system should be setup to test multiple scenarios rapidly and in a controlled environment. Amazon does this. That really should be evidence enough about it’s success.
Distilled is looking for beta testers of Distilled ODN (Optimization Delivery Network). It’s like Optimizely for SEO.
Testing Other Conversions
We should also be testing social shares, amplification and links.
Testing Big
Like we heard from Larry Kim earlier today, small tests lead to small results. Bigger tests will tell you whether something worked or not in a dramatic way.
Machine Learning
He talks about the need to find high-risk pages that could be hit by machine learning and it’s application to user behavior. This is coming and you should fix it now.