The standard core update analysis picks the twenty biggest movers by visibility score. Those are, inevitably, enormous sites — encyclopaedias, retailers, news organisations, platforms with tens of millions of pages.
Then comes the interpretation. Google rewarded X. Google punished Y. The reasoning is plausible and entirely unfalsifiable, because a sample of twenty giant domains cannot tell you what happened to a DR 20 affiliate site in a single niche.
Why the famous names mislead
Large domains move for reasons that do not apply to you. A retailer losing visibility may have changed its faceted navigation. A publisher gaining may have launched a new section.
More importantly, they are outliers by construction. Sorting by absolute visibility change guarantees a list of the largest sites, the same way sorting people by weight guarantees a list of adults.
The question actually worth asking
Not “who won” but “what happened to sites like mine” — same niche, same authority band, same monetisation model.
That is a segmentation question, and it is answerable. Take the sites in one niche, split them by DR band, and compare traffic before the rollout began with traffic after it settled.
| Pattern in your niche | Most likely reading | What to do |
|---|---|---|
| Everyone fell, including the leaders | The whole category was re-rated | Nothing hasty — this is not about your pages |
| Low-DR sites fell, high-DR held | Authority weighting increased here | Links matter more in this niche than they did |
| You fell, comparable sites did not | Something specific to your site | The only case that justifies changing pages |
| Thin pages fell, deep pages rose | Coverage depth was rewarded | Consolidate rather than publish more |
| Everything moved and then returned | Rollout volatility, not a final position | Wait for the rollout to complete before judging |
Timing: the mistake that wastes a quarter
A core update takes weeks to roll out, and rankings move erratically throughout. Measuring on day three produces a number that will not survive to day thirty.
Snapshot the period before the rollout begins, wait until Google confirms it is complete, then snapshot again. Anything measured in between is noise being read as signal.
Search Engine Journal maintains a running history of confirmed algorithm updates which is a reliable way to establish those two dates. Getting the window right matters more than the sophistication of whatever you do next.
What winners usually have in common
Across updates, the sites that hold or gain tend to share characteristics that are unglamorous and difficult to fake.
- Evidence of first-hand experience. Original photographs, measurements, tests, things that went wrong.
- Complete coverage of a defined scope. Not more pages — fewer subjects, handled properly.
- A named, identifiable author with a stated reason to be credible on the topic.
- Pages that answer the query asked, rather than mentioning it on the way to something else.
None of that is new advice, which is itself informative. The updates keep rewarding the same qualities; what changes is how much slack is left for sites that lack them.
Visibility scores are not traffic
Most published winner-and-loser lists are built from visibility indexes — a weighted measure of how often a domain appears for a tracked keyword set.
That is a useful proxy and it is not the same as traffic. A domain can lose visibility across thousands of low-volume terms it never received clicks from, and the headline number will look catastrophic while revenue barely moves.
The reverse happens too. Losing one high-volume query can cost real money while barely registering in a visibility score built from a broad keyword list.
Estimated traffic has its own flaws, but it at least attempts to measure the thing you care about. When the two disagree, prefer the one denominated in visits.
Recovery, honestly
Recovery from a core update is slow and frequently partial. Sites that recover generally do so at the next core update, not before, because that is when the re-evaluation happens.
The corollary is uncomfortable: changes made this week are tested in three months. That argues for making the changes you believe in rather than a scatter of small experiments you cannot attribute.
Two updates beat one
A single update tells you less than people assume. Sites move for all sorts of reasons, and one data point cannot separate a trend from noise.
Comparing across two or three consecutive updates is far more informative. A site that held through all of them is genuinely stable; one that gained and then gave it back was riding volatility rather than earning position.
A repeatable method
- Record the rollout start and completion dates.
- Pull the sites in your niche, split by DR band.
- Compare before and after for each band, not for the niche as a whole.
- Find where your own site sits in that distribution.
- Act only if you moved differently from comparable sites.
Our update view charts before-and-after traffic for every confirmed update, and the explorer filters the same index by niche and DR so you can build that comparison directly. The longer argument for this approach is in reading a Google core update, and the practical question of whether your niche is stable enough to build in sits in validating a niche in an afternoon.
All traffic figures here are third-party estimates. They are dependable for comparing sites against each other through the same window and unreliable as absolute numbers, which is a distinction worth holding onto when a chart looks dramatic.
Done consistently, this turns every update from an anxious week of reading commentary into a short, repeatable measurement — and a decision you can defend either way, including the decision to change nothing.
Frequently asked questions
How long should I wait before judging a core update?
Until Google confirms the rollout is complete, which typically takes several weeks. Rankings move erratically during a rollout, so a measurement taken in the first few days usually describes volatility rather than any final position.
Why did my site drop when my content did not change?
Core updates re-evaluate existing content against what else is available, so a page can lose ground because competitors improved or because the whole category was re-rated. Compare yourself against similar sites before concluding it was something you did.
Can a site recover from a core update?
Yes, though usually at a later core update rather than immediately, and often only partially. That delay means improvements made now are tested months later, which favours substantial changes over a series of small experiments.
Are published winners and losers lists useful at all?
As a signal that an update has landed and is significant, yes. As guidance for a specific site, rarely — they are lists of the largest domains by absolute movement, which is a different population from the one most sites belong to.
The takeaway Ignore the list of famous domains. Segment your own niche by authority band, measure from before the rollout to after it settles, and act only if you moved differently from sites like yours.
