Ahrefs, a widely used search engine optimization (SEO) platform, is facing scrutiny over how it combines public data, proprietary web crawling, and statistical modeling to produce metrics sold through its subscription services.
The company provides estimates and scores covering search volume, website traffic, backlinks, keyword difficulty, and domain authority. Ahrefs operates its own web crawler and maintains large databases. Critics argue that some of the raw information underlying its products is also available through Google tools, open-web repositories, and other outside sources.
Google provides approved advertisers and developers with keyword ideas, approximate monthly search volumes, advertising competition data, and estimated bid ranges. It offers this data through its Keyword Planner and Google Ads API, which returns historical search volume, a competition index, and top-of-page bid ranges for a given keyword. SEO platforms can collect and process that information alongside other datasets before presenting it through proprietary dashboards and metrics.
Ahrefs does not claim that every figure comes directly from Google or represents an exact count. Many of its figures are estimates created from incomplete external signals and statistical models. However, critics say the polished presentation of those estimates can lead users to interpret them as more precise than they are.
Backlink analysis is one area in which Ahrefs has invested heavily in its own infrastructure. The company says it operates a large crawler and maintains a proprietary index of links between websites.
Public alternatives also provide substantial quantities of crawl and link data. Common Crawl, a nonprofit open-data project established in 2007, maintains an open repository of web crawl data totaling more than 10 petabytes, with new crawls published roughly once a month.
Common Crawl also publishes web-graph datasets that map links between hosts and domains. The August–October 2025 host-level release included 468.4 million nodes and 8.0 billion edges, making it a potentially useful source for companies building commercial web-analysis products.
The availability of public data does not mean commercial SEO platforms perform no original work. Collecting, cleaning, storing, deduplicating, and analyzing billions of records requires significant infrastructure and technical expertise. Commercial tools may also combine open datasets with proprietary crawling, historical archives, and internally developed models.
The central question is how clearly platforms distinguish between independently collected information, third-party data, and modeled estimates.
Accuracy concerns are particularly significant in traffic-estimation tools. These tools attempt to estimate private website activity without access to analytics accounts or server logs.
In one case, the publisher Startup Fortune reported that Ahrefs estimated its site received about 3,300 visits in June. The publisher said its actual traffic exceeded one million visitors that month. Based on those figures, the estimate understated traffic by roughly 300 times.
Ahrefs and other external analytics platforms generally cannot view a website’s private first-party records. They must instead rely on signals such as search rankings, estimated click-through rates, observed behavior, and statistical inference. An analysis of Ahrefs’ own internal study, which compared its traffic estimates against Google Search Console data across 1,635 websites, found a median deviation of nearly 50 percent.
Marketers, advertisers, agencies, and publishers commonly use such estimates to compare competitors, prepare reports, and assess commercial opportunities. Large errors can therefore affect negotiations and business decisions when users treat modeled figures as established facts.
Ahrefs’ branded metrics, including Domain Rating and Keyword Difficulty, simplify complex datasets. Keyword Difficulty, for example, scores a keyword from 0 to 100 based primarily on the number of referring domains linking to top-ranking pages. Critics argue that these scores can create an impression of certainty even when the underlying calculations include significant assumptions and uncertainty.
The broader SEO software industry follows a similar model, combining publicly available information, proprietary datasets, and statistical analysis into subscription-based research tools.
Supporters of such platforms say customers are paying not only for exclusive data, but also for convenience, processing speed, filtering, historical records, competitor comparisons, and a more accessible interface.
The dispute is therefore less about whether Ahrefs performs original work and more about whether its pricing, marketing, and presentation adequately reflect the limitations of its estimates.
Greater transparency could include identifying which data is collected independently and which comes from third parties. It could also mean explaining how metrics are calculated, how frequently estimates are tested against first-party data, and what margins of error users can expect.
Ahrefs’ tools can provide useful directional analysis when estimates are reasonably accurate. Critics say customers can avoid treating individual figures or proprietary scores as definitive by comparing them against first-party analytics or additional data sources.
Sources
- Google for Developers. “Keyword Planning overview.” Google Ads API documentation.
- Common Crawl Foundation. “About Common Crawl.”
- Common Crawl Foundation. “Host- and Domain-Level Web Graphs, August, September, and October 2025.”
- Startup Fortune. “Hey Ahrefs, Maybe It’s Time You Retire Your SEO and Traffic Tools.”
- Tech Business News. “How Incorrect Ahrefs Traffic Estimates Can Be Based on Search Console Data.”
- Ahrefs. “Keyword Difficulty.” Ahrefs SEO glossary.