Problem
Ecommerce traffic dropped after a Google core update: how to fix it
A core update is a change in how results are assessed rather than a penalty applied to your site. Nothing was removed and there is nothing to appeal. The useful response is to establish that the update is genuinely the cause, find which page types and query types lost, and then improve those pages on the terms the update rewards. Reacting quickly with broad changes usually makes the next reading harder rather than the traffic better.
By CartKernel · Last reviewed 2026-09-07
Does this look familiar?
- Organic clicks fell over a few days and settled at a lower level
- The decline lines up with the announced start of a broad core update
- Category and guide pages lost more than product pages, or the reverse
- Average position slipped across many queries rather than a few
- Impressions held while clicks fell, meaning positions moved down the page
- Competitor pages that now rank read as more thorough or more clearly written
Causes, ranked
Why it happens, most common first
Check them in this order. The first two account for most cases we open.
- most common
Content is thinner than what now ranks above it
Category pages with a heading and a grid, or product pages carrying supplier text, offer less than pages that answer the questions a buyer has. Core updates repeatedly shift results toward pages that resolve the query completely, and stores tend to be light on that kind of content.
- most common
The site shows little evidence of first hand expertise
Assessment favours content that demonstrates real experience with the products: how they fit, how they compare, what goes wrong, who they suit. Pages assembled from specifications alone read as a catalogue rather than as advice, and lose to sites that show they have handled the products.
- common
Search intent for the query moved
An update can change what kind of page is judged appropriate for a query. Where results shift toward guides, comparisons or a different page type, your product or category page may lose position without becoming worse. This shows as losses concentrated in particular query patterns.
- common
A large volume of low value pages weighs on the site
Thin filter pages, near duplicate variants, tag archives and mass generated content contribute little and are assessed alongside everything else. Sites carrying a lot of it often see broad, site wide movement rather than page specific losses.
- occasional
Page experience makes the content harder to use
Heavy interstitials, layout shifting, slow interaction and content buried under promotional blocks reduce how well a page serves the visitor. This rarely causes a drop by itself, but it separates two otherwise comparable pages.
- occasional
The drop is not the update at all
Seasonality, a tracking change, a site release, a redirect error or a lost integration can coincide with an update. Attributing everything to the update leads to months of unnecessary rewriting while the real cause sits in a change log.
The fix
In this order
Each step is something you can do today. Do them in sequence; skipping ahead is how a review fails twice.
Prevent it next time
- Build category and guide content that answers the question rather than listing products alone
- Keep a written release log so a traffic change can be matched against site changes
- Review thin and near duplicate pages quarterly and consolidate rather than accumulate
- Report organic performance by page type so a shift is visible early
Confirm the timing and rule out other causes
Compare the daily clicks curve with the announced rollout dates, then check your release log, analytics configuration and Search Console for errors on the same days. Look at year over year data so a seasonal dip is not mistaken for an update.
Find exactly what lost
Export query and page data for the four weeks before and after, and compare clicks and average position by page type, by query type and by whether the query is brand or generic. The pattern in that comparison is the whole diagnosis.
Read the results that replaced you
For the ten queries with the largest losses, look at what now ranks and describe honestly what those pages do that yours does not. Do this as a written note per query. Vague impressions produce vague changes.
Improve the affected pages rather than the whole site
Rewrite the specific pages that lost, starting with the highest value ones. Add what the query needs: sizing and fit detail, comparison against alternatives, compatibility, care and returns information, and answers to questions your support team actually receives.
Show experience the reader can verify
Add original photography, measurements you took, notes on how products perform in use, named authorship where a person genuinely wrote it, and clear information about the business behind the store. Remove claims you cannot substantiate.
Reduce the pages that add nothing
Consolidate overlapping collections, exclude filter combinations that were never worth indexing, and either improve or remove thin pages that carry no traffic. Fewer, stronger pages is the direction updates have consistently rewarded.
Fix the experience issues you already know about
Address layout shift, slow interaction and intrusive overlays on the affected page types. These are worth doing regardless, and they remove an obvious difference between your page and the one that outranks it.
Give the work time and measure at the next assessment
Changes are reflected as pages are recrawled and reassessed, and meaningful movement often waits for a subsequent update. Track by page type monthly rather than checking rankings daily, and resist making a second round of changes before the first can be read.
When to get help
It is worth getting help when the loss is large enough to matter commercially and the pattern is not obvious from your own data, or when previous attempts at recovery have not moved anything. An outside review is mostly useful for the honest comparison against what now ranks, which is difficult to do about your own pages. Be wary of any plan that promises a specific recovery date or a return to previous positions, because neither can be committed to.
Free. We reply within one business day.
Can a site recover from a core update?
Yes, and recovery normally shows up as pages are reassessed, often around a later update rather than immediately after changes go live. Recovery is not automatic and it is not guaranteed by activity alone. It follows from the pages genuinely becoming a better answer than the ones that replaced them.
Should I wait for the next update before changing anything?
Diagnose immediately and improve the pages you are confident about, but do not make sweeping site wide changes on a guess. Work that is right for the reader is worth doing whenever you do it. Work done as a reaction to a chart usually has to be undone later.
Does removing old content help after a core update?
Removing pages that serve no purpose and attract no traffic can help by concentrating attention on the rest of the site. Removing pages that still earn clicks does not. Review each page against its traffic and its role before deleting anything, and prefer improving or consolidating over deleting.
How do I tell an update from a technical problem?
A technical problem usually affects a specific address pattern, produces errors in coverage reports, or coincides exactly with a release. An update tends to move average position across many queries with no errors anywhere. Check the release log and coverage reports before accepting the update as the explanation.
Services that fix it for good
Related problems, answers and terms
Stop the leak.
A free Growth Analysis finds what is broken and ranks it by the revenue it costs you.