Google runs several core updates a year. In 2026 so far there have been two: the March 2026 core update, which started on 27 March and took about 12 days, and the May 2026 core update, which started on 21 May and finished in early June. Each one leaves some sites with pages that lost rankings and a long list of articles that might need work.
Search Console tells you which pages lost traffic. It can't tell you which of them let readers down, and those are the ones a refresh is most likely to help. This post shows how to combine the two and put the list in order.
The short answer
Start with the articles that lost the most search clicks and that readers also rate poorly. Wait until the update has finished and a week has passed, find the pages with the biggest sustained drops in Search Console, then check what readers said on those pages. Pages with many lost clicks and a low share of "Yes, this helped" come first. Pages that dropped but that readers still rate well need a closer look before you change them.
Wait until the update has finished
Rankings move a lot while an update rolls out, and early changes don't always last. Google's own advice is to:
- Check on the Google Search Status Dashboard that the update has finished rolling out, and note its start and end dates.
- Wait at least a full week after it ends before you analyse anything.
- Compare that week with a week before the update started.
Google also says small drops in position usually need no drastic action, and warns against quick fixes. Treat the first week as a time to gather data, not to rewrite.
Find the pages that lost the most
In Search Console, open the Performance report and compare the two periods:
-
Sort pages by the change in clicks
Look at absolute losses first. A page that went from 2,000 clicks to 1,200 matters more than one that went from 20 to 5, even though the second lost a bigger share.
-
Check position as well as clicks
A page that lost clicks but kept its average position may be losing them to an AI summary, which is a different problem from a ranking drop.
-
Open the queries for each page
Note which searches it dropped for. A guide that lost one important query needs different work from one that slipped a little on all of them.
Tip: Keep the list short. Twenty pages you can look at properly are worth more than two hundred you skim.
Add what readers said about those pages
Google asks site owners to judge their pages with questions such as "Will someone reading your content leave feeling like they've had a satisfying experience?". On a page that asks readers whether it helped, you already have their answer to a version of that question.
For each page on your list, note three things from your reader feedback over the months before the update:
- The share of readers who answered Yes, if the page has at least 10 answers.
- The reasons readers picked most often, such as "Out of date" or "Didn't answer my question".
- Any comments that point to something you can check, such as a wrong price or a broken step.
Reader answers are not a ranking signal, and Google never sees them. They tell you which pages people found lacking, which is a good guide to where a refresh can do some good.
Put the list in order
Put each page in a row of this table. Work from the top down:
| Priority | Search clicks lost | Readers said | What to do |
|---|---|---|---|
| 1 | Many | Mostly No | Refresh it now. Readers had already told you it fell short, and it lost the most. |
| 2 | Few | Mostly No | Refresh it next. The loss is small, but readers still struggle with it. |
| 3 | Many | Too few answers | Check it against your best pages on the same topic and read any comments. Ask readers before you rewrite. |
| 4 | Many | Mostly Yes | Look before you change it. Readers liked it, so compare it with the pages that now rank above it. |
Pages that lost little and that readers rate well can wait. Changing a page readers like, only because its rank moved, risks making it worse for them.
Decide what kind of update each page needs
The reasons readers pick tell you what kind of work a page needs, which also tells you how long it will take:
| Readers mostly pick | Likely work | Effort |
|---|---|---|
| Out of date | Check facts, prices, versions and screenshots | Hours |
| Steps were unclear | Rewrite the steps that comments point to | Hours to a day |
| Didn't answer my question | Compare the page with the queries it lost and cover what is missing | A day or more |
| Too long or hard to follow | Restructure: answer first, detail after | A day or more |
| Something else | Read the comments before you plan anything | Varies |
Example
Example: a home improvement site after a core update
Say you run a home improvement site with 400 how-to guides, each ending with "Did this guide solve your problem?" and "What would make it better?". A week after a core update finishes, Search Console shows 30 guides with large drops in clicks.
You check each one against your results for the three months before the update. Eight have mostly No answers, and on six of them readers mostly picked "Out of date": product names and prices that changed last year. Those six go to the top, and an editor updates them in a week. Your guide to tiling a bathroom also dropped, but 85% of its 60 readers answered Yes. You leave its text alone, look at the pages now ranking above it, and add the cost table they have and it lacks.
For each refreshed guide, you compare reader answers over the same number of weeks before and after the change, as well as its clicks in Search Console.
Check whether the refresh worked
Google says some changes show in search within days, while others take months, and some only after the next core update. There is no guarantee that any change brings traffic back.
Reader answers come back sooner. If the share of Yes answers on a refreshed page rises over the following weeks, the page is better for the people who read it, whatever its ranking does. If it doesn't, read the new comments before you try again.
Where Asklet fits
Asklet asks the same short question at the end of every article from one snippet, and keeps the path each answer came from. You can pick any date range up to 52 weeks to see what readers said before an update, then compare the same number of weeks after a refresh. It shows counts until a page has 10 responses, so a page with three answers doesn't jump up your list by accident.
It doesn't connect to Search Console or know your rankings: you line the two up yourself, page by page. Readers answer without signing in, and the widget sets no cookies and uses no tracking identifiers.
Your next steps
-
Confirm the update has finished
Check the Google Search Status Dashboard, then wait a full week.
-
List the pages with the biggest sustained click losses
Compare a week after the update with a week before it in Search Console.
-
Add each page's reader results from before the update
The share of Yes answers, the top reasons and any checkable comments.
-
Sort the list with the priority table
Many lost clicks and mostly No answers first.
-
Match the work to the reasons readers gave
Fix facts first: they are quick and readers notice them.
-
Compare reader answers before and after each refresh
Use the same number of weeks, then watch Search Console over the following months.
Try these steps on your own site. Asklet adds the questions to every page with one snippet. Free plan, no card needed. Compare the plans.
Start freeCommon questions
How long after a core update should I wait before changing pages?
Google recommends waiting until the update has finished rolling out and then at least a full week before you analyse the change. Rewriting pages while rankings are still moving makes it hard to tell what had an effect.
Will updating an article bring its rankings back?
Not necessarily. Google says there is no guarantee, and that some changes take months to show or only show after a later core update. Update pages because readers need it, and expect the search effect to be slow.
Should I just change the date on an article to make it look fresh?
No. Google's helpful content guidance lists changing dates to make pages seem fresh, when the content hasn't really changed, as a warning sign. Readers who pick "Out of date" want the facts updated, not the date.
What if a page that dropped has no reader feedback yet?
Start collecting it now. Meanwhile, compare the page with the pages that now rank above it and with your own best pages on the topic, and fix any facts you can check yourself.
Related
Sources
More from the blog
- Privacy and law EU cookie consent changes: where the Digital Omnibus stands in October 2026 9 min read
- Privacy and law Session replay and pixel lawsuits: what website owners should know 9 min read
- Privacy and law How to make a feedback widget accessible, and what the European Accessibility Act asks 9 min read