Rebecca Berbel talks with Jason Barnard about predictive SEO using big data.
This is an extremely interesting and insightful episode where Rebecca Berbel and Jason Barnard discuss predictive SEO: where it came from, where it is now, and where it is going.
Rebecca also cleared some misunderstandings between predicting in SEO and predictive SEO, ranking factors and features, and the explainable and unexplainable elements that contribute to the machine’s predictions of ranking on the SERPs.
As always in SEO, there are loads and loads of “it depends”, so it doesn’t look like predictive SEO will change that quirk in our industry 🙂 Rebecca and Jason squeeze dozens of knowledge nuggets, tons of machine learning insights, and masses of SEO ranking analysis in this episode 😉 Also, watch out for the answer to this pragmatic question: How can we present predictive SEO data to clients?
What you’ll learn from Rebecca Berbel
- 00:00 Rebecca Berbel with Jason Barnard
- 01:53 Rebecca Berbel’s Brand SERP and her event Knowledge Panels
- 05:03 What is predicting in SEO and what is predictive SEO?
- 07:55 Contexts where “it depends” for ranking
- 09:51 Straight forward machine learning algorithms versus the black box
- 12:37 Thinking of factors as ML features
- 15:48 Why do you need to clean data? (garbage in, garbage out)
- 21:30 The three types of people in SEO
- 24:55 Explainability with Shapley (game theory)
- 29:20 Different sections of your site don’t all behave the same way
- 33:01 Examples of negative SEO factors evaluated by OnCrawl’s predictive model
- 36:36 Examples of positive SEO factors evaluated by OnCrawl’s predictive model
- 38:30 How to win Google’s game with data
- 40:33 How can we present predictive SEO to clients?
- 41:59 What output does predictive SEO provide?
- 44:11 The next steps in predictive SEO
This episode was recorded live on video December 7th 2021
Recorded live at Kalicube Tuesdays (Digital Marketing Livestream Event Series). Watch the video now >>