Split Test SEO Experiments (Kevin Indig and Jason Barnard)

Kevin Indig talks with Jason Barnard about split test SEO experiments.

Kevin Indig is SEO Director at Shopify and is the creator of the Growth Memo newsletter and Tech Bound podcast. In the past, he worked at G2 and Atlassian as head of SEO and has helped companies like eBay, Eventbrite, Samsung, Pinterest, and many others grow their organic traffic.

To take your SEO activities to an exceptional level, you need to remove the guesswork. But how would you do that? A rigorous and scientific approach called statistical SEO split testing is the key.

In this fascinating and insightful talk, Kevin Indig and Jason Barnard lead us through the theory and practicalities behind evaluating SEO strategies using split testing.

Here, they discuss how split testing works both on templated and non templated sites,how to plan, the pointers that matter in the process, how long the process takes to finally evaluate the results and the external factors that might occur in the middle of split testing.

Kevin is delightful, funny and he is super obsessed with this topic: he’s definitely the geek you would want to call when it comes to running analytical SEO experiments 😉

P.S. Kevin also answered some live chat questions from SEO split testing experts. Watch out !!

What you’ll learn from Kevin Indig

00:00 Kevin Indig and Jason Barnard
01:21 Kevin Indig’s Brand SERP in America and Australia
03:22 What are Split Test SEO Experiments?
05:33 Kevin’s insights into Google core algorithms
08:20 Types of SEO testing for aggregator sites
11:36 Split testing on non templated websites
13:23 What matters more than the number of control pages
15:52 Every SEO test is a quasi experiment
19:09 Pointers for DIY split testing
21:47 Subtle factors that can make experiments go sour
24:37 Coverage of Kevin’s split testing environments
26:07 Transferring Google Ads split testing to SEO strategy
28:02 The cool down period before running another test
30:32 How to deal with external factors in the midst of split testin
33:14 Differences between neural networks and Google Causal Impact for analysing results