Why comparison sites matter for AI recommendations

Comparison pages are worth your attention because they appear in the research behind AI recommendations. Ahrefs found that brands higher on third-party lists were also more likely to appear in responses. Our published analysis ranked comparison sites first under both influence tests. That makes relevant lists worth investigating, without treating a placement as a certain recommendation.

What does the research actually show?

In the Ahrefs study, best-of blog lists accounted for 43.8% of source-link page types. The researchers examined ChatGPT answers to 750 early-stage questions about software, products and agencies. Those are the questions the finding describes, rather than AI search as a whole.

We re-judged 6,517 answers for our published analysis, keeping the model and judging rules behind every citation verdict. Comparison and ranking sites came first under both tests. We judged their role in answers; we did not test whether buying a listing caused more recommendations.

Which kind of comparison page are you trying to join?

A recommendation blog and a product category page are different kinds of listing. Ahrefs separates blog lists from platforms such as G2, where options are ranked in another format. Keep that distinction when deciding which kind of placement you want.

On G2, your software’s features determine which categories it can join. The product must meet all a category’s requirements, so the decision goes beyond who uses it.

How do you add your product to G2?

Search G2 for your product before submitting anything. If it is missing, send the product form for the research team to check eligibility and category. Once approved, claim the profile to customise its details and add your branding.

Check that the product qualifies first. G2 excludes consumer-only products and those still in alpha or beta, the early testing stages.

Support a category request with evidence that the product meets its requirements. G2 suggests recorded demonstrations, blog content or presentations, and its research team makes the final decision. Submitting the request is only part of that process.

What makes a comparison worth reading?

A useful comparison helps the reader understand the choice. Google asks review writers to show their own experience and explain how the options differ, including benefits and drawbacks. Use those checks to assess a comparison page’s quality, without treating them as a promise of AI citations.

A “best” label needs a reason. Google says recommendations for best overall or best for a purpose should explain why, backed by first-hand evidence. Look for that support when assessing the listing.

Should you publish a list that puts your product first?

Agency owner Wil Reynolds draws a line between being cited and convincing a buyer. He argues that putting yourself first on your own list does not build trust. Read this as his marketing judgement, rather than a measured rule about AI rankings.

Ahrefs describes work beyond publishing a list that names itself the winner. It focuses on improving its product, sharing original research and taking part where its audience is. Those are useful alternatives for product teams to consider.