After a core update: confirm the impact, assess honestly, recover patiently

What a core update is and is not

Google describes core updates plainly: “Several times a year, Google makes significant, broad changes to our search algorithms and systems.” They are announced on Google's list of ranking updates and appear on the Search Status Dashboard with a start and end date.

The key word is broad. “These changes are broad in nature, and don't target specific sites or individual web pages.” Google explains it with a restaurant list: if a friend asks for your top 20 restaurants years after you first wrote the list, new places have opened and you have new experiences, so the list changes. In Google's words, “The list will change, and restaurants that move down aren't necessarily "bad"; there are just other restaurants that make your top 20.”

That gives you three things a core update is not:

• Not a manual action. Nobody at Google reviewed your site. There is no message in Search Console and no reconsideration request to file.
• Not a spam update. Spam updates target practices that break Google's spam policies. A core update reassesses how helpful and reliable content is across the whole web.
• Not a list of things to remove. Google does not publish what changed in a specific update, so there is no checklist of "banned" page elements to strip out.

The full picture of ranking systems, spam policies and manual actions is covered in the lesson "Ranking systems, core updates and spam policies". This lesson is about the practical part: what you do in the weeks and months after an update.

What Google officially says

The table below separates what Google actually tells you from the things people often assume it said.

Step 1: confirm the update actually affected you

Many drops blamed on a core update have nothing to do with it. Prove the connection before you spend months fixing the wrong thing.

Get the dates from the Search Status Dashboard

The Search Status Dashboard shows the current status of four parts of Google Search: crawling, indexing, ranking and serving, with an incident history. Ranking updates, including core updates, are listed there with the date the rollout started and the date it finished. Google's first instruction is: “Confirm that the core update has finished rolling out.” Write both dates down; every comparison you make depends on them.

Wait, then compare the right weeks

Rankings move during the rollout, so early numbers are unreliable. Google recommends waiting a week after the rollout ends, then: “After a week, try comparing this week with a week before the core update started rolling out; this way you'll be able to better pinpoint what exactly has changed.”

In practice, keep the periods equal: 7 days against 7 days, or 28 against 28, with the same weekdays. Avoid comparisons where one period contains a public holiday or a sale. Then check the last 16 months in the Performance report to see whether the same dip happens every year at this time.

Check that the timing really matches

Look at daily data. If clicks started falling two weeks before the rollout began, the update is not the cause. If the drop appears inside the rollout window, the link is plausible but still not proven, because a release on your side could have shipped the same week. Check your own deployment and content calendar for that period.

Rule out the other causes

Google's traffic-drop guide lists technical issues, security issues, spam issues and manual actions, seasonality and changing interests, site moves, and reporting glitches as separate causes. Check the Page indexing, Crawl stats, Security issues and Manual actions reports, and confirm your analytics tracking did not change. The full method is in the lesson "Why traffic dropped: a diagnostic method".

Step 2: size and segment the change

Google draws a clear line between small and large drops, with its own examples:

• Small drop. “Small drop in position (dropping from position 2 to 4): There's no need to take drastic action (in fact, we recommend avoiding making changes to content that's already performing well).” A move like this can still cost a lot of clicks, because the top positions get most of them, but it is not a sign that the page is poor.
• Large drop. “Large drop in position (dropping from position 4 to 29): Do a deeper assessment.” This is the case the rest of this lesson is about.

Next, find where the loss sits. Google puts it this way: “For example, one important factor is finding out if the drop happened across the whole site, a group of pages, or even just one very important page in your site.” Segment in this order:

1. Search type. Web, Images, Video and News are separate in the Performance report. Google: “This can help you understand whether the drop you've seen happened in Web Search, Google Images, Video mode, or the News tab.”
2. Section or template. Use page filters with regex such as ^https://example\.com/blog/ or /product/ to compare blog posts, product pages, category pages and tools separately.
3. Query groups. Brand against non-brand, informational against commercial. A drop only in "how to" queries points somewhere very different from a drop in product queries.
4. Country and device. A loss in one market can come from a local competitor or a translation problem.
5. Individual pages. Export the Pages table for both periods, sort by clicks difference and list the biggest losers and the biggest gainers. You need both lists for Step 3.

The filters, regex and comparison modes are explained in the lesson "Performance report: filters, regex, comparisons".

When it is not you

A core update can coincide with changes that have nothing to do with your quality. Check these before you start rewriting:

• Clicks fell, impressions did not. Google notes that after a small position change “you might see a noticeable drop in traffic without a big change in impressions.” And more generally: “If your impressions remain the same but your clicks drop, you might not be generating the best page title and snippet that you could, and so users don't understand the content of your page, or perhaps other sites had a more appealing rich result.”
• The results page changed shape. Search your lost queries. If AI Overviews, video, forum results or shopping blocks now sit above the organic results, your position may be similar but your visibility is not.
• A better page arrived. Google's own example: “there may be other pages on the web that are doing a better job of helping the searcher.” And: “If there's more deserving content, it will continue to rank well with our systems.” Look at who replaced you and be honest about whether they deserve it.
• Demand moved. Interest in a topic can fall for reasons unrelated to Google. Check your top queries in Google Trends.

None of these is fixed by a content overhaul. Snippet issues need better titles and descriptions; a changed results page may need a different format; a fall in demand needs new topics, not edits.

Step 3: assess your content honestly

For a sustained, large drop, Google asks you to read its self-assessment questions and check whether your site “overall (not just individual pages)” delivers helpful, reliable, people-first content. Two instructions shape the method: “Take a close look at your site as a whole, and try to be objective.” and “Evaluate the pages that were most impacted.”

1. Build a sample

Take the biggest losers from Step 2, split by template or section, and add pages that held or gained. Twenty to fifty pages is usually enough to see a pattern on a medium site. Without the gainers you cannot tell what the update rewarded on your own site.

2. Compare with what ranks now

For each lost page, search its main query and open the pages that now rank above it. Note what they offer that yours does not: first-hand testing, original data, clearer structure, a more current answer, a named expert, a better format for the intent. Google's helpful content page asks: “Does the content provide substantial value when compared to other pages in search results?”

3. Answer the questions, page by page

Score each page against Google's questions with a simple scale such as yes, partly, no. Start with the content questions, for example “Does the content provide original information, reporting, research, or analysis?”, then the expertise questions, then Who, How and Why: who wrote it, how it was made, and why it exists. On E-E-A-T, Google is clear that “Of these aspects, trust is most important.” The questions themselves are worked through in the lessons "Helpful content in practice" and "E-E-A-T: the four pillars"; author pages and bylines are covered in "Authors, About and editorial policy".

Google also suggests looking at the type of search, not just the page: “What pages were most impacted and for what types of searches?” A page can be fine for one intent and weak for another.

4. Get an outside view

You know your site too well to judge it neutrally. Google's advice: “Beyond asking yourself these questions, consider having others you trust but who are unaffiliated with your site provide an honest assessment.” Give the reviewer the same pages and the same scale, and compare their scores with yours. Large disagreements are where your blind spots are.

5. Look for patterns

Single weak pages rarely explain a site-wide drop. Look at the scores by template and section: an affiliate section with no evidence of testing, a blog full of rewritten summaries, location pages that differ only by the city name, articles with no author, advice that is years out of date. These patterns are what you fix.

Step 4: prioritise improvements

You will find more problems than you can fix at once. Rank them:

1. Site-level patterns first. A fix to a template or an editorial process improves hundreds of pages; a rewrite improves one.
2. Then by value. Lost clicks multiplied by how much those pages matter to the business: a product category is worth more than an old news post.
3. Then by how fixable it is. Adding a real author and sources is quick; producing original research is not. Mix quick and slow work so something ships every week.
4. Leave winners alone. Pages that held or gained are your reference, not your to-do list.

Google's description of good changes is useful as a filter: “Consider how you can improve your content in meaningful ways. For example, it could be that rewriting or restructuring your content makes it easier for your audience to read and navigate the page.” For the mechanics of refreshing, merging and pruning, use the lesson "Content audit: prune, merge, refresh".

Deletion sits at the bottom of the list. “Deleting content is a last resort, and only to be considered if you think the content can't be salvaged.” Google adds that wanting to delete whole sections tells you something: “In fact, if you're considering deleting entire sections of your site, that's likely a sign those sections were created for search engines first, and not people.” In that case, Google says removing the unhelpful content can help the good content perform better.

What not to do

Panic produces most of the damage after a core update. Avoid these:

• ✕Quick fixesRemoving a widget, an FAQ block or ads because someone said it hurts SEO. Google explicitly warns against “quick fix” changes.
• ✕Mass deletion or noindex without analysisRemoving hundreds of URLs because they have low traffic loses pages that may have been fine and does not prove anything about the rest.
• ✕Padding to a word countGoogle asks whether you write to a word count and answers its own question: “(No, we don't.)”
• ✕Fake freshnessChanging the date without changing the page. Google lists it as a warning sign: “Are you changing the date of pages to make them seem fresh when the content has not substantially changed?”
• ✕Bulk adding or removing content "for freshness"Google asks whether you do this hoping it will make the site “seem "fresh?" (No, it won't)”.
• ✕Rewriting pages that held upGoogle recommends “avoiding making changes to content that's already performing well”.
• ✕Link panicBuying links or filing a large disavow in reaction to a core update. A core update is not a link penalty.
• ✕Changing everything at once, unloggedIf you ship twenty changes in a week and do not record them, you will never know which helped.

Realistic recovery timelines

This is the part clients most want a date for, and Google gives none. Its words: “If you've made improvements, it may take time to see an effect in Search results: some changes can take effect in a few days, but it could take several months for our systems to learn and confirm that the site as a whole is now producing helpful, reliable, people-first content in the long term.”

Then the sentence to quote in every client conversation: “If it's been a few months and you still haven't seen any effect, that could mean waiting until the next core update.” There is a second route too: “However, you don't necessarily have to wait for a major core update to see the effect of your improvements.” Google runs smaller core updates as well: “These updates are not announced because they aren't widely noticeable, but they are another way that your content can see a rise in position (if you've made improvements).”

For measuring, the traffic-drop guide suggests a rhythm: “In general, you'll likely want to wait a few weeks to analyze your site in Search Console again to see if your efforts had a beneficial effect on ranking position.” And always keep the caveat in view: “Keep in mind that there's no guarantee that changes you make to your website will result in noticeable impact in search results, and that positions in Google Search results aren't static or fixed in place.”

Documenting the work

Six months from now nobody will remember what changed on which date. Keep a change log from day one, in a shared spreadsheet or your ticket system:

Add the update's start and end dates, and your largest releases, as annotations in your reporting tool so every chart shows them. Save the Search Console exports for the baseline weeks: Search Console keeps 16 months of data, and you will want the original numbers later. Store the assessment scores too, so the next review can compare like with like.

Communicating with stakeholders and clients

A drop after a core update is stressful for whoever pays for the site. Clear, early communication prevents the panic changes described above.

• Explain what happened. A broad ranking change, not a penalty. Google's restaurant analogy works well with non-specialists.
• Show the confirmed impact, segmented. "Non-brand clicks to review pages fell after the rollout; product and category pages held" is more useful than one total line.
• Share the plan and the reasons. What you will change, in what order, and why those pages.
• Be honest about timing. Quote Google: improvements can take months, sometimes until the next core update, and there is no guarantee.
• Report on a fixed rhythm. Monthly against the same baseline. Separate effort (pages improved, sections reworked) from outcome (clicks, impressions, positions) so progress is visible before the traffic returns.
• Don't promise a date or a percentage. Nobody outside Google can know it.

Client-facing reporting is covered in more depth in the lesson "SEO reporting".

Common mistakes

• ✕Analysing during the rolloutRankings are still moving; wait a week after the end date.
• ✕Unequal comparison periodsA 7-day period against a 10-day period, or a holiday week against a normal one.
• ✕Blaming the update without checking timingThe drop began before the rollout, or matches your own release.
• ✕Skipping technical checksIndexing errors, a stray noindex or broken tracking look like a ranking drop.
• ✕Only looking at losersWithout the gainers you cannot see what the update rewarded.
• ✕Judging single pages, missing the patternThe problem usually sits in a template or an editorial habit.
• ✕Grading your own workNo outside reviewer, so every page scores "good".
• ✕Promising recovery by a dateGoogle gives no timeline and no guarantee.

FAQ

Sources

• Google Search Central — Google Search's core updates and your website
• Google Search Central — Debugging drops in Google Search traffic
• Google Search Central — Creating helpful, reliable, people-first content
• Google Search Status Dashboard

Quotes checked against the official pages on 30 October 2026.