Impact analysis of Google's May core update
My analysis of the 2020 May Core Update
“Don’t chase the algo” is a popular SEO response to anyone who dares to post about the update on Twitter. “If you have to make changes after you’ve been impacted, it’s too late” is another one.
So, why am I writing about updates, then? Because, while it’s true that you shouldn’t optimize for algorithms, it’s important to understand updates. They deliver helpful information about what Google rewards (at that point in time) and where the journey is going. A simple “provide the best content and UX possible” simply doesn’t cut it.
On May 4th (Star Wars day!), Danny Sullivan posted the following on Twitter:
A follow-up Tweet states: “As is typical with these updates, it will typically take about one to two weeks to fully roll out.” Looking at rank trackers, this doesn’t seem to be true.
Algoroo shows the biggest fluctuation on May 6th:
According to CognitiveSEO, there was no sign of a roll-out in Italy. Maybe it’s still coming.
SERPmetrics shows the same kind of fluctuation:
Mozcast shows rainy weather:
RankRanger confirms that this update has impacted mobile and desktop rankings to the same degree:
If we want to take Accuranker as a gage, this update is big but not as big as the one in October 2018 or October 2017 (to be fair, SERP volatility can be caused by different kinds of updates):
I also encourage you to check out the country-based SEPR volatility on SERPWoo.
According to SEMrush, categories like “Arts & Entertainment” have had much more SERP volatility than others and top stories are now shown a lot more often, which aligns with some observations that news publishers gained more visibility from the update:
In fact, the biggest volatility was observed in the news sector by SEMrush.
As always, Danny also referred to this post on the Google blog about core updates.
To take it away upfront, this update seems to have been massive. About 3-5x a year, Google updates its core algorithm next to the hundreds or thousands of little updates it makes all the time. The common notion is that core updates refresh ranking ML algorithms with lessons from the previous months. The machine becomes smarter.
Loyal Tech Bound readers know that since episode #89, I’ve been writing about “ML algo updates” and that my thesis is that all core updates since Jan 16 are connected with each other:
We’ve seen substantial algorithm updates since January 2016:
January 2016 (https://www.seroundtable.com/google-core-ranking-21460.html)
December 2017 (https://www.seroundtable.com/google-hanukkah-update-24928.html)
The biggest losers/winners
Rankranger, SEMrush, and Searchmetrics published a list of winners and losers on Search Engine Land. Lily Ray from PathInteractive published an analysis of 550 domains based on the Sistrix visibility index.
But I also personally track a couple hundred sites on desktop and mobile across many verticals - my personal index - to get an idea of algorithm impact. Besides winner/loser domains of the actual update, I like to track a fixed set of domains to understand algorithm updates.
(I didn’t include Linkedin, which had a technical issue that deindexed their whole site)
Let’s look at some examples of impact.
9gag.com got hit really hard on Desktop…
… and gained on mobile
Investing.com saw a steep increase:
Imdb.com dropped after already having seen a drop in traffic on 4/16:
Tiktok.com dropped on desktop after having gained on 4/18…
… and kept gaining on mobile.
Even though it’s still early, I was able to spot a couple of patterns in the update:
It rolled out within 2 days
It rolled out pretty much at the same time in most countries
In many cases, a change in user intent led to different SERP features being shown, which means steady rankings but changing traffic for impacted sites
I didn’t spot a pattern that YMYL sites very affected more than others
Some sites saw huge changes, other subtle ones
Don’t miss the deep dive into Spotify on the paid Friday episode this week!