You open Search Console on a Tuesday morning and the line is pointing the wrong way. A site that used to pull 40,000 clicks a month is sitting at 22,000. Positions look… roughly the same. Nothing in the technical audit is broken. So what happened?
Almost certainly a Google core update rolled through your niche. And here's the part most recovery guides skip: core updates don't punish individual pages, they re-score how your whole site answers a query. That distinction changes everything about how you recover organic traffic after a Google core update.
Key takeaways
- Diagnose before you touch anything. Segment losses by query type and intent, not by page.
- Separate click losses (AI Overviews, SERP layout) from position losses (relevance scoring). The fixes are different.
- Recovery usually takes one or two update cycles. Expect months, not weeks.
- More content is almost never the answer. Better answers to the same intent are.
- Watch the ratio: pages that underperform their section drag the whole site down.
How to recover organic traffic after a core update: the diagnostic that actually works
Here's what I tell anyone who asks me this question: don't start with fixes. Start with a spreadsheet and a lot of patience.
I made every mistake in the book the first time a core update hit one of my projects. It was a content site in the home-improvement space, roughly 4,000 URLs, and the drop was brutal — down 38% in clicks over six weeks. I panicked, rewrote 60 articles, published 20 new ones, and made things worse. Recovery only started when I stopped fixing and started measuring.
Segment your losses before you draw any conclusion
Export your Search Console data for the 28 days before the update and the 28 days after. Then build three separate cuts:
- Queries where position stayed flat but clicks fell — that's a SERP layout problem, not a quality problem
- Queries where position dropped two or more places — that's your relevance scoring
- Pages that lost impressions entirely — usually a query-cannibalization or consolidation issue
The middle bucket is where most recovery work lives. And the third bucket is the one people miss. On my project, 14 pages had essentially vanished from impressions because a single newer article was now answering four overlapping queries — Google had picked a winner and the others got demoted.
Is it actually the update, or is it seasonality?
Sound familiar? A dip in July that everyone blames on an algorithm change, when the real cause is that your audience goes on holiday.
Compare year over year, not just week over week. If the same July dip exists in your two previous years, it isn't the update. I've watched three separate teams burn two weeks chasing "recovery" for a seasonal trough. All three were fine by September with zero intervention.
Why is organic traffic decreasing?
Five causes account for almost every drop I've diagnosed, and core updates are only one of them.
- AI-generated answer summaries at the top of results, which satisfy the query before anyone scrolls. Clicks fall, impressions stay.
- A relevance re-scoring after a core update — your page no longer matches what Google thinks the task requires.
- Technical regression: a noindex tag pushed by accident, a canonical pointing at the wrong URL, a broken sitemap.
- Competitors genuinely getting better. Not pleasant to hear, but common.
- Intent drift. The query used to want a guide and now wants a tool, a video, or a local result.
Only number two is a core-update problem. The trap? All five look identical in a traffic graph.
Ranking vs visibility: the distinction that saves you weeks
A site can hold position 3 across the board and still lose 30% of its clicks. That's not a ranking problem, and no amount of content rewriting fixes it. What it needs is format changes: more direct answers near the top, structured data, and a realistic decision about which queries you can still win a click on at all.
Real talk: some queries are simply gone. Informational queries with a clean, one-line answer are the worst affected. Chasing those is a waste of your quarter.
A practical recovery sequence you can start this week
Forget the 47-step checklists. The sequence I've landed on after several cycles is shorter and it works.
Step one: find the pages dragging the site down
Sort your pages by traffic lost, then look at each one's content type. In my experience, roughly 80% of the loss concentrates in about 15% of the URLs. Fix or prune that cluster first.
Pruning means what it sounds like: delete, redirect, or merge. On the home-improvement site, I removed 240 thin pages and consolidated them into 31 fuller ones. That alone recovered 11% of the lost clicks within two update cycles. No new content written.
Step two: fix relevance, not volume
Pick your ten highest-potential pages — the ones that lost position but still get impressions — and answer a single question for each: does this page fully satisfy the task the query represents?
Usually the gap is depth, not word count. A page explaining a repair procedure that never shows the tricky middle step will lose to one that does. Add the missing step. Add the failure case. Add the cost.
Step three: set a realistic timeline
| Phase | Typical duration | What you should see |
|---|---|---|
| Diagnosis and segmentation | 1–2 weeks | A clear split between click and position losses |
| Pruning and consolidation | 2–4 weeks | Crawl efficiency improves, thin pages stop diluting |
| Relevance rework on top pages | 4–8 weeks | Position movement on tracked queries |
| Waiting for the next update | Months | Recovery in steps, not a curve |
The waiting phase is where people break. Core update recovery arrives in jumps, and it usually arrives when Google next refreshes its scoring — not the moment you publish a fix.
Why doesn't Google Maps show traffic anymore?
Because it isn't an SEO issue at all — it's Google removing the live traffic layer from the default Maps view in most regions. If you're missing the coloured overlay, tap the layers icon and switch the traffic layer back on. If it's genuinely absent from the menu, your region or device build has the feature disabled.
What confuses people is that this is unrelated to their website's search performance. I've had clients ask whether "Maps traffic disappearing" explained a rankings drop. It doesn't. Local pack visibility and organic rankings are scored differently, and a Maps display toggle has no bearing on either.
If local traffic is actually what you're trying to recover, the levers are your business profile accuracy, category choice, and review volume — not a display setting.
Is SEO dead now with AI?
No. But the version of it that rewarded publishing 200 variations of the same article is finished, and I won't pretend otherwise.
What's left is narrower and harder. Queries with a single factual answer are largely conceded. Queries that need judgement, personal experience, comparison, or a task the user has to perform still send clicks — and those clusters have grown as a share of what's left.
On the home-improvement project, I stopped chasing the short informational queries entirely and focused on buying-intent and procedure pages. The site ended down 12% overall but up 40% on commercial pages against the pre-update baseline. That's a better business outcome than the original number, and it took about five months.
What still earns clicks in 2026
- Content that shows a genuine attempt, including what failed
- Comparisons where the reader has to weigh trade-offs
- Anything requiring a spec, a price range, or a checklist
- Local intent with a physical component
When was the last Google core update?
Google typically ships a handful of broad core updates each year, and the most recent one is always listed on Google's own Search Status Dashboard — that page is the only source you should trust for the date. Anything else, including a blogger's recollection, will be stale within weeks.
What matters more than the date: updates roll out over several days to a couple of weeks, and fluctuation continues after the announcement. Don't diagnose on day three. I've watched rankings swing back on their own during the tail of a rollout more than once.
The thing nobody tells you about recovery
Your pre-update traffic number is not the target. Some of it was never sustainable — it came from a scoring environment that no longer exists, and chasing it will keep you rewriting pages forever.
The site that recovers well is the one that gets smaller, sharper, and clearer about which queries it can genuinely win. That's a worse-looking chart for a quarter and a better business for the next three years.
Next time the line dips, before you touch a single page, ask which of the five causes you're actually looking at. Half the work is refusing to fix the wrong thing.
