By Moberstein

Posted by Moberstein

I’m completely fascinated by Google’s Discover Feed. Besides the fact that it serves highly-relevant content, it also seems beyond the reach of being gamed. In a way, it almost seems beyond the reach of pure SEO (which makes it downright tantalizing to me).

It all made me want to understand what makes the feed tick.

So I did what any sensible person would do. I spent the better part of two months running all sorts of queries in all sorts of different ways to see how it impacted my Discover Feed.

Here are my ramblings.

My approach to analyzing Google’s Discover Feed

Let me explain what I did and how I did it, to both give you a better understanding of this analysis and point out its gaping limitations.

For five days a week, and over the course of two months, I executed all sorts of user behavior aimed at influencing my Discover Feed.

I ran queries on specific topics on mobile, searched for other topics on desktop… clicked on results… didn’t click on results… went directly to sites and clicked… went directly to sites and didn’t click anything, etc.

In other words, I wanted to see how Google reacted to my various behaviors. I wanted to see if one behavior influenced what showed in my Discover Feed more than other behaviors.

To do this, I searched for things I would generally never search for, went to sites I would normally never visit, and limited my normal search behavior at times so as not to influence the feed.

For example, I hate celebrity news and gossip with a passion, so I went to people.com every day (outside of the weekends) and scrolled through the site without clicking a thing. I then recorded if related material (i.e. celebrity nonsense) ended up in my Discover Feed the next day.

I recorded all of my various “web behaviors” in the same way. I would execute a given behavior (e.g. search for things related to a specific topic on mobile, but without clicking any results) and record what happened in my Discover Feed as time went on.

Here’s a breakdown of the various behaviors I executed along with the topics associated with each behavior. (For the record, each behavior corresponds to a single topic or website so I could determine the impact of that behavior on my Discover Feed.)

Allow me to quickly elaborate on the behaviors above:

• These are all topics/sites that I am in no way interested or involved in (particularly self-help car repair).
• When I clicked a YouTube video, I watched the entire thing (I mean, I didn’t actually watch half the time, but Google doesn’t know that… or do they?)
• When I visited a site, I did scroll through the content and stay on the page for a bit.
• A search for a “segment of a topic already existing in Discover Feed” means that the overall topic was something that regularly appeared in my feed (in this case, NFL …read more

  Source:: Moz Blog