White-label links & AI search for agencies
Get a tailored proposal in 1 business day — no contracts.
James AllsoppSemrush found substantially more organic keywords than Ahrefs across every website and country we tested — between 6.2× and 2,192× more, with a median gap of 50×. Google Search Console independently confirmed the sites had far broader search visibility than Ahrefs' smallest counts suggested. Ahrefs remains excellent for backlink research; for keyword breadth, our data favours Semrush.
Ahrefs and Semrush can give remarkably different answers to a seemingly simple question:
How many keywords does this website rank for?
For years, marketers have accepted some variation between competing SEO tools. Each platform maintains its own databases, collection infrastructure and estimation models, so identical numbers would actually be surprising.
But the differences we started seeing in 2026 went far beyond ordinary variation.
For one website, Semrush reported 2,192 organic keywords in the US.
Ahrefs reported one.
For another, the figures were 1,160 versus three.
That raised an obvious question.
Rather than judging the platforms by whichever dashboard showed the larger number, we compared them with first-party Google Search Console data from four websites to which we have verified access.
Across those four websites, we analysed:
and multiple country-matched Ahrefs and Semrush datasets.
The results were surprisingly consistent.
The interesting part wasn't simply that Semrush showed more keywords. It was that Google's own data repeatedly showed a much broader search footprint than the tiny numbers we were seeing in Ahrefs.
— James Allsopp, iNet VenturesOur conclusion is narrower than saying one platform is universally “better”.
But for organic keyword coverage, Semrush was considerably more comprehensive across every website we tested.
We carried out the study in August 2026.
Four websites were selected from different niches:
We have deliberately anonymised the domains in this article.
The objective of the research is to compare SEO data providers — not publish commercially useful keyword opportunities from websites we operate.
For each site, we collected:
We then compared the breadth of search visibility shown by each source.
This follows the same underlying investigation detailed in our earlier study, Is Ahrefs Missing Organic Keywords? We Tested It Against Google. read our investigation into Ahrefs' missing organic keywords
The difference here is that we're concentrating specifically on Ahrefs versus Semrush.
One trap when comparing Ahrefs and Semrush is assuming their headline worldwide keyword totals are necessarily constructed in identical ways.
They aren't always.
Ahrefs' API documentation, for example, states that its Organic Keywords metric represents the number of keywords where a target ranks within the Top 100 organic search results. In its “all locations” methodology, the same keyword appearing across different countries can still be counted once.
Semrush also defines its Organic Positions report around keywords where a domain appears within Google's Top 100 results in the selected regional database.
We therefore concentrated heavily on country-matched comparisons.
US Semrush versus US Ahrefs.
UK Semrush versus UK Ahrefs.
Canada versus Canada.
And so on.
That gives us a much cleaner experiment.
Here were eight of our major country-level comparisons:
| Website | Market | Ahrefs | Semrush | Difference |
|---|---|---|---|---|
| Site A | US | 3 | 1,160 | 386.7× |
| Site A | UK | 0 | 503 | — |
| Site B | US | 31 | 192 | 6.2× |
| Site B | UK | 32 | 270 | 8.4× |
| Site C | US | 1 | 2,192 | 2,192× |
| Site C | UK | 2 | 438 | 219× |
| Site D | US | 14 | 418 | 29.9× |
| Site D | UK | 1 | 50 | 50× |
Semrush reported more organic keywords in every comparison.
If we exclude the comparison where Ahrefs returned zero — because a multiplier cannot sensibly be calculated — Semrush reported between:
6.2× and 2,192× more organic keywords.
The median difference was:
50×
That's considerably larger than the sort of database disagreement we'd normally expect.
But a larger number isn't automatically a more accurate one.
That's where Google Search Console becomes important.
Flexible link building plans with allocation across clients. Track everything via your live dashboard.
Across the four sites, Search Console exposed 4,000 query rows from the 28-day test periods.
Their average reported position distribution looked like this:
| GSC reported position | Query rows |
|---|---|
| Positions 1–3 | 183 |
| Positions 4–10 | 543 |
| Positions 11–20 | 379 |
| Positions 21–50 | 1,660 |
| Positions 51–100 | 1,233 |
| Position 100+ | 2 |
| Total | 4,000 |
That means:
Those aren't estimates from Ahrefs or Semrush.
They're queries visible in the Google Search Console accounts of the websites themselves.
There are important caveats around GSC position data, which we'll discuss later, but the overall footprint is unmistakable.
Google was showing these websites across a large query universe.
This is the easiest result.
🏆 Winner: SemrushSemrush found more keywords on all four sites and in every major regional database we compared.
Site C was the standout example:
Site A:
Site D:
Even the least dramatic example, Site B, produced:
That's still an 8.4× difference.
If your primary objective is competitor keyword research, this matters considerably.
A platform showing 30 keywords instead of several hundred can cause you to underestimate:
If I'm analysing competitors, I would rather start with the dataset that gives me the broader search landscape and then filter it down. You can't investigate an opportunity the tool never showed you in the first place.
— James Allsopp, iNet VenturesThis is more important than simply asking which platform showed the larger number.
Google Search Console demonstrated that the sites had far more query visibility than the smallest Ahrefs figures suggested.
Take Site C.
GSC exposed:
Ahrefs meanwhile reported:
one US organic keyword.
Semrush reported:
2,192.
Does GSC prove that exactly 2,192 current US rankings existed at one moment?
No.
But does it make a universe of one US ranking keyword look implausibly narrow?
Yes.
The same thing happened elsewhere.
Site D had:
336 GSC queries averaging position 10 or better
while Ahrefs reported just:
14 US keywords
and only 28 unique keywords worldwide in the snapshot we tested.
Semrush showed:
418 US keywords.
The Semrush number is therefore much more consistent with the breadth of visibility Google itself was recording.
We need to be careful with the word accurate.
There are at least two different ways a keyword database can be accurate:
When the tool says a website ranks for a query, is that ranking genuine?
How many of the genuine rankings does the tool actually discover?
These are not the same thing.
Consider two hypothetical tools analysing a site with 1,000 real rankings.
Tool A
Finds 40.
All 40 are correct.
Tool B
Finds 800.
760 are correct and 40 are stale.
Tool A could have exceptionally high precision.
But Tool B is substantially more useful for understanding the overall search footprint.
Our testing suggests that Ahrefs' issue is much more likely to be recall/coverage than widespread inaccurate rankings.
In fact, when we investigated Ahrefs keywords from one of the sites, all five we checked were also represented in its recent Search Console data.
The tool wasn't necessarily inventing rankings.
It just seemed to be finding far fewer of them.
This isn't a claim Ahrefs entirely disputes.
Ahrefs explains that organic traffic is estimated by first finding keywords for which a target ranks, then modelling traffic using ranking position, search volume and estimated CTR. It explicitly tells users to treat those numbers as estimates rather than site analytics.
Its API defines Organic Keywords as the total number of keywords the target ranks for in the Top 100 organic results.
The interesting issue isn't whether Ahrefs has every Google query.
No third-party SEO platform does.
It's how large the blind spot appears to have become on certain domains.
We explored that question separately in our original Ahrefs missing organic keywords investigation. see the full Ahrefs investigation and methodology
This was an obvious possibility.
If Semrush reports 2,000 keywords and Ahrefs reports 20, perhaps Semrush is simply carrying thousands of dead rankings.
So we inspected Semrush rankings on our test domains.
What we found were large numbers of highly relevant query-to-page relationships.
Semrush was associating topic-specific queries with corresponding articles on the sites — exactly what you'd expect from genuine organic ranking data.
More importantly, we found examples where:
We have chosen not to publish those keywords because doing so would reveal commercially useful information from our sites.
That prevents us from publishing a public false-positive percentage.
But it does mean the additional Semrush universe isn't simply a random collection of nonsensical keywords.
Interestingly, Semrush provides an Organic Traffic Insights product specifically designed to combine its own keyword data with Google Search Console metrics.
Its documentation explains that users can switch between Semrush keywords and GSC queries to understand positions, impressions, clicks and landing-page performance together.
That's an important acknowledgement of a basic truth:
Third-party SEO databases and first-party Google data answer different questions.
Neither should automatically replace the other.
Our results are not isolated.
In July 2026, GEO Toolbox published an independent comparison using three websites and checking Ahrefs and Semrush against Google Search Console.
Its results across three sites:
| Site size | Ahrefs keywords | Semrush keywords |
|---|---|---|
| Small / new | 35 | ~1,100 |
| Medium | 367 | 2,089 |
| Large | 26,835 | 80,700 |
Semrush found more ranking keywords on all three.
That study comes with an important caveat: the author notes that geographic scopes weren't always perfectly aligned, so the exact ratios shouldn't be treated as controlled laboratory measurements.
Still, the direction of the result is remarkably similar to ours.
It also found something useful for avoiding an overly simplistic conclusion:
Ahrefs produced closer traffic estimates on some of those sites.
That's entirely plausible.
Thousands of low-volume long-tail rankings can add enormous keyword breadth while contributing little actual traffic.
This is why our study asks:
Which is more accurate for organic keywords?
not:
Which SEO platform is universally more accurate?
Links placed into aged, indexed content with ~5 day turnaround. No new posts needed.
In May 2026, a peer-reviewed study published in the Journal of Information Science examined Ahrefs, Semrush, Serpstat and Ubersuggest.
Researchers Cristina I. Font-Julian, Carlos Lopezosa, Enrique Orduña-Malea and Isidro F. Aguillo compared organic keywords, search volumes and URL-level traffic across major academic publishing websites.
Their central finding was that SEO-tool selection can materially affect the results of a study because the datasets differ so substantially.
The authors concluded:
“The differences between the tools are great enough” to produce different comparative conclusions.
They also found that restricting analysis to Top-10 rankings produced more reliable comparisons than relying on total organic-keyword counts.
That recommendation is particularly interesting in the context of our own data.
Across our four sites, Google Search Console exposed:
726 queries averaging reported Top-10 positions.
So our findings aren't being driven exclusively by thousands of position-90 terms.
There was substantial first-page visibility too.
Here's where things become particularly interesting.
A 2022 comparison covered by Search Engine Land examined 51 websites using data from Google Search Console, Ahrefs, Semrush and Moz.
In that sample, Ahrefs had the largest third-party keyword database for:
50 of 51 websites.
Ahrefs was reported to find roughly twice as many keywords as Semrush on average.
That's almost the reverse of what we're observing in 2026.
Different sites and methodologies mean we should not claim that this proves Ahrefs' database has objectively declined.
But it does raise an obvious question:
One important event occurred in September 2025.
On September 10, 2025, Google stopped supporting the long-used &num=100 behaviour that enabled SEO tools and other automated systems to retrieve approximately 100 Google results in one response.
Ahrefs describes the event in its own Google algorithm/change history as affecting the entire industry, noting that positions outside the Top 10 suddenly became more difficult and expensive to obtain.
Ahrefs publicly acknowledged that it would initially see a reduction in keyword counts because deeper SERPs had become harder to retrieve.
That event is highly relevant to any 2026 conversation about organic ranking databases.
But it should not be misrepresented.
Ahrefs worked around the change.
Its subsequent updates said it had restored Top-100 retrieval for Rank Tracker, meaning it would be incorrect to argue that Ahrefs simply stopped collecting rankings beyond page one permanently.
The more interesting question is subtler:
Did the economics and methodology of collecting deep long-tail SERPs change how different platforms build their broader domain-ranking databases?
We cannot answer that from outside either company.
The internal architectures are proprietary.
But the 2025 event demonstrably made Top-100 SERP collection harder across the industry.
And our 2026 results show the two platforms now exposing dramatically different keyword universes on some domains.
The relationship is worth investigating, but it would be irresponsible to claim direct causation.
One debate following Google's num=100 change was whether anyone really needs positions 21–100 anyway.
Some SEO professionals argued the Top 20 contains nearly all actionable data.
Independent SEO consultant Brodie Clark took a different view:
“Personally, I'd ideally like to have the top 100 still.”
His reasoning was that movement below the Top 20 can still help diagnose SEO issues and identify developing trends.
We agree.
A URL moving:
#87 → #54 → #31 → #17
might not generate traffic until late in that journey.
But an SEO analysing the domain would absolutely want to know that the page is gaining visibility.
That's particularly important when using rank tracker tools to monitor emerging performance.
For competitor research, missing that journey makes competitors look weaker than they really are.
There's another reason we chose current 2026 Search Console data rather than relying on historical GSC benchmarks.
After Google's change, many website owners saw sudden changes in Search Console impressions and average positions.
Brodie Clark reported examples where impressions declined by as much as 47%, arguing that automated rank trackers had likely been contributing artificial Search Console impressions when repeatedly requesting deep result sets.
That makes pre/post-September-2025 Search Console comparisons messy.
Our study avoids that particular problem:
all of our GSC data comes from August 2026.
We are therefore comparing current third-party datasets with current post-change Search Console behaviour.
Google's search results have also changed dramatically in another way.
AI Overviews mean that a Search Console “position” doesn't necessarily correspond neatly with a classic blue-link position.
Ahrefs' current organic-keyword API even recognises AI Overview and AI Overview sitelinks as SERP feature types, illustrating how much broader modern SERPs have become.
Ahrefs itself analysed one million SERPs in 2025 and found AI Overviews appearing on more than a quarter of the keywords in that sample.
That introduces another reason we are careful with the wording of our GSC results.
We say:
“GSC queries averaging a reported Top-10 position.”
We do not claim:
“Every query represented a traditional blue-link Top-10 ranking.”
The modern SERP is more complicated than that.
However, that complication doesn't invalidate the underlying coverage comparison.
Google is still recording visibility for these sites across hundreds or thousands of queries.
That remains inconsistent with some of the exceptionally tiny Ahrefs keyword footprints we observed.
Based on this experiment:
Semrush
At least when the objective is finding the breadth of organic keywords a competitor may be visible for.
The advantage isn't merely that Semrush returned more rows.
It's that Google's first-party data repeatedly supported the existence of a much broader search footprint than Ahrefs was showing.
For more traditional SEO analysis tools, this is a useful reminder that “more conservative” does not automatically mean “more accurate.”
You need to know what type of accuracy matters for the task.
For competitor keyword discovery, missing rankings is a major weakness.
For predicting traffic, the calculation may be very different.
This result doesn't mean we'd remove Ahrefs from our stack.
Ahrefs remains exceptionally useful for:
In fact, this study reinforces the argument for combining providers rather than becoming dependent on one.
Our own Bulk Domain Intelligence system follows exactly that philosophy, combining Ahrefs, Moz and Majestic metrics so users can compare DR, DA, Trust Flow, Citation Flow, organic traffic and more than 20 additional signals together.
The same idea powers our Backlink Analyzer, which cross-references data from Ahrefs, Moz and Majestic rather than pretending a single proprietary metric can provide a complete picture.
Different datasets have different strengths.
We analyze competitor backlinks and secure the same high-quality links for your site.
This may be the most useful takeaway from the experiment.
If Ahrefs says:
14 organic keywords
it does not mean Google knows only 14 queries for that website.
If Semrush says:
418 organic keywords
that doesn't mean Google knows exactly 418 either.
These are:
third-party observations of Google
not:
Google's master ranking database
The distinction is similar to authority metrics.
Domain Rating, Domain Authority, Trust Flow and Authority Score are useful measurements — but they are generated by particular providers.
They are not Google ranking scores.
That's why we prefer to cross-reference metrics using tools such as Bulk Domain Intelligence rather than allowing a single number to make the decision.
The same principle applies to organic keywords.
For organic keyword coverage in our August 2026 sample:
Semrush won clearly.
Across four websites:
Semrush reported more organic keywords in every country comparison.
Where Ahrefs returned at least one keyword, Semrush reported:
6.2× to 2,192× more.
The median difference was:
50×.
Meanwhile Google Search Console exposed:
across the same four sites.
Those first-party Google datasets repeatedly suggested the websites' search footprints were much broader than Ahrefs' organic-keyword counts implied.
The fact that:
all point toward significant database differences makes this more than an isolated dashboard anomaly.
Semrush gave us a substantially broader picture of competitor keyword visibility.
Nothing replaces first-party impressions and clicks when you actually own the property.
Keyword breadth is one metric. Backlinks, traffic estimates, historical analysis and workflow quality are different questions.
No.
The most important conclusion from this research is:
An SEO platform's organic keyword count measures that platform's view of Google — not Google's complete view of the website.
And those views can currently be remarkably different.
I'd still happily use both tools, but for different jobs. Based on this research, if I wanted to uncover the widest possible set of competitor rankings tomorrow, I'd start in Semrush and use Ahrefs as another source rather than assuming Ahrefs' keyword total was definitive.
— James Allsopp, iNet VenturesThere are several important limitations.
⚠️ Before you rely on these numbers, keep the following in mind:
Therefore our conclusion is deliberately specific:
In our test, Semrush offered substantially broader organic keyword coverage, and that broader footprint was more consistent with the scale of visibility shown by Google Search Console.
We are not claiming that every Semrush position is correct.
We are not claiming Ahrefs is inaccurate across every metric.
And we're not claiming these exact ratios apply across the entire web.
In our August 2026 study, Semrush provided substantially broader organic keyword coverage across all four websites tested.
Country-matched comparisons showed Semrush reporting between 6.2× and 2,192× more keywords where Ahrefs had at least one result, with a median difference of 50×.
Google Search Console independently showed a much broader query footprint than Ahrefs' smallest keyword totals suggested.
Ahrefs and Semrush maintain different keyword databases and use different collection, processing and refresh methodologies.
Both platforms report organic rankings using their own observed keyword universe rather than Google's complete query dataset.
Google's September 2025 removal of the num=100 behaviour also made retrieving deep SERPs more resource-intensive across the SEO industry.
Semrush says its Organic Positions report shows keywords for which a domain has pages ranking within Google's Top 100 results in the selected regional database.
Ahrefs' current API documentation defines its Organic Keywords metric as the number of keywords for which a target ranks in the Top 100 organic search results.
Google's 2025 changes temporarily made deep SERP collection more difficult, but Ahrefs subsequently worked to restore Top-100 collection.
For websites you own, Search Console is the key first-party source for Google impressions and clicks.
However, its average-position metric shouldn't be treated as a conventional rank tracker because results can be averaged across dates, locations, devices and SERP experiences.
Ahrefs does not have access to your private Search Console ranking universe when analysing arbitrary competitors.
It works from the keywords and SERPs represented in its own database.
Search Console, meanwhile, reports first-party visibility recorded for your own property.
The two systems therefore should not be expected to match.
Based on this particular study, we would currently start with Semrush for competitor keyword discovery because its coverage was dramatically broader.
For important analysis, however, we recommend checking more than one provider rather than treating either database as complete.
Not necessarily.
A smaller dataset can still contain highly accurate observations.
Our results suggest Ahrefs' issue on the tested sites was primarily coverage: the platform appeared to detect a much smaller subset of the broader Google footprint.
Yes, and for serious competitor research that may be the best approach.
You can also use iNet Ventures' collection of free SEO tools for additional research, including keyword planning, backlink analysis and domain intelligence.
If you've arrived here because your own Ahrefs dashboard suddenly looks unusually empty, read our companion investigation:
Is Ahrefs Missing Organic Keywords? We Tested It Against Google.
That study goes deeper into the apparent Ahrefs coverage problem itself, Google's 2025 num=100 change and why modern AI-powered SERPs make traditional keyword-count metrics increasingly complicated.
Read: Is Ahrefs Missing Organic Keywords? We Tested It Against Google
For a broader overview of choosing platforms for different workflows, see our guide to the best SEO tools for different budgets.
And if you're analysing large numbers of potential competitors, publishers or link prospects, Bulk Domain Intelligence lets you compare Ahrefs, Moz and Majestic metrics side-by-side rather than basing the decision on a single provider.
James is the founder and CEO of iNet Ventures, bringing over two decades of experience in SEO, link building, and online growth. He has helped online agencies scale through white-label solutions and has overseen the delivery of 25,000+ links, combining proven strategies with AI-driven workflows to drive efficient, long-term results.
Want Link Building Without The Hassle?
Our team handles outreach, placements, and reporting — 100% white-label, fully managed.
Start A CampaignRelated Articles
Continue exploring our latest insights and strategies for digital marketing success.
![How to Increase X Followers - 30 Tips and Tricks [2026]](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F90630dz3%2Fproduction%2Fdd58619f91e7cf66ffddbc41dea4435f6458221f-1672x941.png%3Fw%3D1200&w=3840&q=75)
