Google announced the May 2020 core update on 4 May, and a lot of SEO people were not happy about the timing. Plenty of businesses are already dealing with demand that has collapsed or exploded in the space of six weeks, and a broad ranking update on top of that makes it very hard to tell what is causing what.
I have some sympathy with the complaint. I also don’t really expect Google to stop running its search engine because the rest of us are having a difficult spring.
What we know
Not much, as usual. Google confirmed the update on its Search Liaison Twitter account, said it would take a week or two to roll out fully, and pointed to the guidance it published last August about what to do if you lose rankings after a core update. The short version of that guidance is that there’s nothing specific to fix, and you should focus on content quality and the questions in the blog post about expertise and trustworthiness.
The rank tracking tools all registered it as a big one, comparable to the larger updates of the last couple of years. The early commentary has the usual mix of winners and losers lists, with some sectors apparently moving more than others. I’d treat those lists with caution. Search behaviour itself has shifted a lot since March, so some of the movement people are attributing to the update is probably just the world changing underneath the results.
Why this one is harder to read
Normally when traffic drops after a core update, you can line up the dates, check Search Console, and get a reasonable idea of whether it was rankings or demand. At the moment that’s much harder:
- Demand has moved around. Searches for anything to do with home, garden, cooking and fitness have gone up. Searches for travel, events and office clothing have fallen off a cliff.
- Year-on-year comparisons are mostly useless. Last May was a normal month; this one isn’t.
- Competitors are behaving oddly. Some have paused paid search, some have stopped trading online, some have discounted heavily. All of that changes click-through behaviour on the organic results too.
The way I’m trying to separate it out is to look at average position and impressions in Search Console for a stable set of queries, rather than at clicks or revenue. If position has dropped across many queries since 4 May, that’s probably the update. If position is steady and clicks are down, that’s more likely demand.
What I’d actually do
If you’ve lost visibility, give it a couple of weeks before making changes. Rollouts take time, and some sites that drop early come back partway once things settle. Then look at the pages that lost the most and ask uncomfortable questions about them. Are they genuinely the best answer for the query? Is it obvious who wrote them and why anyone should trust them? Would you be comfortable with someone from Google’s quality rater programme looking at them?
For ecommerce sites, I’d look hard at category pages and thin product pages. Two lines of manufacturer copy and a price doesn’t give Google much reason to rank a page over a competitor with the same product. I’ve written before about selling craftsmanship online, and most of what makes a product page convincing to a human seems to help with search too.
If you’ve gained, don’t assume it’ll last. I’ve seen sites gain on one core update and hand it back on the next.
Meanwhile, something more concrete
The more useful Google announcement this month came the day after the update. On 5 May Google introduced Web Vitals, a small set of performance metrics it wants everyone to focus on: Largest Contentful Paint for loading, First Input Delay for interactivity, and Cumulative Layout Shift for visual stability. They’re measured from real users in Chrome rather than from a lab test.
Google hasn’t said these will become a ranking factor, but the direction is fairly obvious, and unlike a core update you can measure them and fix them. Layout shift in particular is a problem on a lot of ecommerce sites, where images and promotional banners load late and push everything down the page just as someone’s about to tap. I’d much rather spend time on that than on guessing what changed on 4 May.