Can You Trust Your Keyword Ranking Reports Right Now
Keyword ranking reports have become less reliable since September 2026, because Google has been returning swapped results to many of the automated checks that rank trackers depend on. For businesses that judge SEO for B2B companies tracking their rankings by a weekly position report, that matters. A sudden drop in the report may not be a drop at all.
The problem was first reported by Search Engine Roundtable and has since been measured by several tracking vendors. A small test on 20 October shows how it plays out in practice, using searches where this website already appears.
What Has Changed in Google Results for Rank Trackers
Rank trackers work by running searches automatically and recording where each website appears. Since mid September, a share of those automated searches have come back with complete, normal looking results pages in which the usual websites have been replaced by others.
The rank tracking company Nozzle published a study called Poisoned Google SERPs. It found 8.4% of the results pages it judged since 13 September had been swapped, against 0.17% before. The worst days were 27 to 29 September, when the rate peaked at 28.6% across the 52 countries the study covers. It also found that a tracked website was missing from the top 10 on 60.2% of the swapped pages, compared with 2.7% normally. Most swapped pages did not last, with the usual websites back within three days in 68.9% of cases.
Other vendors have seen the same thing at different rates. AccuRanker uses a different method and has reported much lower rates in October, which shows how much the measured scale depends on how each tool checks its data. SISTRIX has explained on its data collection page that it now publishes only results that pass quality checks and collects failed keywords again.
What a Small Test Showed
Five keywords were checked on 20 October 2026 to see whether the problem showed up in practice. Each is a search where this website appears in Search Console with an average position between 4 and 9. Every keyword went through DataForSEO, a data provider many rank trackers are built on. Each was also checked through a second provider, SerpApi. Two keywords were checked through DataForSEO a second time a few minutes later.
| Keyword | Search Console average position (20 September to 17 October) | DataForSEO first check | DataForSEO second check | SerpApi check |
|---|---|---|---|---|
| healthcare web design | 4.0 | 1 | Not run | 1 |
| woocommerce agency | 8.8 | 4 | Not in top 10 | Not in top 10 |
| google ads consultant | 7.8 | Not in top 10 | Not in top 10 | 5 |
| wordpress hosting | 7.9 | Not in top 10 | Not run | Not in top 10 |
| microsoft ads management | 6.5 | Not in top 10 | Not run | Not in top 10 |
For two keywords the two providers returned the same results in the same order. For the two keywords checked twice, the picture changed from one check to the next.
The clearest example was “woocommerce agency”. The first DataForSEO check placed our guide at position 4, in a results page that shared only one website with the SerpApi check. A few minutes later the same search through DataForSEO returned a page sharing eight of nine results with SerpApi. This time the guide was not in the top 10.
“Google ads consultant” went the other way. SerpApi showed the guide on this website at position 5. Two DataForSEO checks a few minutes apart left it out of the top 10 and did not share a single page with the SerpApi results.
Five keywords is a small sample and proves nothing on its own. It does show how a weekly report built from one automated check could record a rise or a fall that never happened for real searchers.
Why Search Console Will Not Match a Single Check
Search Console comes directly from Google and records what happened in real searches, so it is the most trustworthy source of ranking data you have. It measures something different from a rank tracker, though. The two should not be expected to agree.
Google’s explanation of Search Console metrics says average position is the topmost position of your website in each search, averaged across every search where it appeared. That average blends devices, locations and personalised results, while a rank tracker records one search from one place at one moment.
Search Console also has limits. It cannot show competitors, it groups results in ways that change over time and its data arrives a couple of days late. Its filters and comparisons are what make it useful, so it is worth learning them before relying on it for decisions.
How to Read Ranking Reports Until This Settles
Until tracking data settles, ranking reports are most useful as a guide to trends across many keywords rather than a verdict on any single one. A few habits keep decisions grounded in what searchers see.
- Check Search Console before reacting to a sudden drop.
- Look at trends across groups of related keywords.
- Ask your tool provider how it handles suspect results.
- Rewrite a page because of one bad week in a report.
- Judge an SEO campaign on single keyword positions.
- Compare September and October rankings without context.
Clicks, impressions and enquiries tell you more than positions on their own, because they show whether visibility is turning into business. Our guide to measuring SEO success covers which figures give a fair picture over time.
When a drop does look real in Search Console as well, the usual checks still apply. Our guide to what to do when your Google rankings drop walks through them in order, starting with technical causes before content.
Setting Up a Search Console Check for Your Key Terms
A simple Search Console view can act as a weekly sense check on any ranking report. It takes a few minutes to set up and gives a figure that comes from real searches rather than automated ones.
In the Performance report, filter by the query or page you care about and compare the last seven days with the seven days before. The Performance report guide also explains how to filter by country and device, which matters for a UK business whose customers search mostly from one market.
Clicks and impressions are the most useful pair to watch. If impressions hold steady while a rank tracker reports a big fall, the fall is unlikely to be real. If impressions and clicks both drop in the same week, the change is worth investigating properly.
Saving the comparison as a regular routine makes it easy to repeat. Over a few months it also builds a record of how the website really performs, which is useful when tracking data is questioned again.
What This Means for SEO Reporting
Ranking reports for September and October 2026 need reading with care, because they overlap both the swapped results and a period of wider ranking movement. Our piece on ranking volatility in 2026 covers why positions can swing even when nothing on the website has changed.
Reports that lead with Search Console clicks, impressions and conversions are far less exposed to the problem. Keyword positions still have a place, but as supporting detail rather than the headline figure a decision rests on.
If a report shows a keyword falling out of the top 10, search for it in a private browser window and check Search Console for the same week. If both disagree with the report, the data is the likely problem.
It is also reasonable to ask providers for their error rates. Several vendors now filter or recollect suspect results. Knowing how often that happens helps you judge how much weight a single report can carry.
Indexing problems can look like ranking drops in a report as well. Our guide to indexing issues explains how to tell a page that has left the index apart from one that has simply moved.
Keeping Decisions Based on Real Searches
Google has not commented publicly on the swapped results. It is not clear how long they will continue. Commentators believe the aim is to deter automated scraping more widely, with rank trackers caught up along the way.
Whatever the cause, the safest response is to base decisions on data from real searches. Search Console, analytics and the enquiries a website generates show what is happening far more reliably than any single automated check. Rank tracking then becomes a useful extra rather than the measure everything else is judged against.
FAQs
Why has my rank tracker shown a sudden drop that recovered a few days later?
Since September 2026 Google has returned swapped results to many automated rank checks, so a tracker can record a page that does not match what searchers see. Nozzle found most affected pages returned to normal within a few days, which is why a short dip followed by a recovery is often a data problem rather than a real ranking change.
Is Search Console more accurate than a rank tracker?
Search Console comes directly from Google and reflects real searches, so it is the more reliable record of how a website performs. It reports an average position across every search, device and location, which means it will rarely match a single check from a rank tracker.
Should we stop using rank tracking tools?
Rank trackers are still useful for watching competitors and spotting trends across groups of keywords. They are less reliable for single keywords on single days at the moment, so treat any sudden change as a prompt to check Search Console before acting.
Has Google explained why rank trackers are getting different results?
Google has not made a public statement about it. Commentators and tool vendors believe the measures are aimed at automated scraping more widely, with rank trackers caught up in the effect.
Co-Founder at Priority Pixels
Paul leads on development and technical SEO at Priority Pixels, bringing over 20 years of experience in web and IT. He specialises in building fast, scalable WordPress websites and shaping SEO strategies that deliver long-term results. He’s also a driving force behind the agency’s push into accessibility and AI-driven optimisation.
Nathan Yendle | 20/10/2026
