hi, i’m amr — and the headline number on review sites right now looks like a funeral notice: G2’s organic traffic fell 84.5% between 2024 and the end of 2025. the second number, sitting right next to it in the same data, says the opposite.
the traffic collapse that’s real
SE Ranking’s analysis found organic traffic from 2024 to the end of 2025 fell 84.5% for G2, 89% for Capterra, and 92% for TrustRadius. those aren’t seasonal dips. the cause is where buyers moved: 51% of B2B software buyers now start their research in an AI chatbot, up from 29% a year earlier, and 69% of buyers say they chose a different vendor than originally planned based on AI guidance. AI chatbots are now the number one influence on the shortlist, ahead of review sites.
the citation concentration that’s just as real
the same study found 88% of review citations in Google’s AI Overviews still go to just five review platforms. the humans stopped clicking through; the machines kept reading. that’s the paradox: review sites are getting less direct traffic and more indirect influence, because AI engines pull category and comparison answers heavily from the same handful of platforms, and a vendor’s review profile is now part of what an AI says about it before any buyer ever visits the site.
the consolidation that changes the playing field
G2 acquired Capterra, Software Advice, and GetApp from Gartner in a deal that closed on February 5, 2026, reportedly for around $110 million. the combined group holds nearly 6 million verified reviews and reaches more than 200 million software buyers annually, and analysts estimate it now controls roughly 53-58% of global software-review influence by several independent measures. one company now sits behind most of the review surface buyers — and AI engines — actually read.
what review traffic still converts at
buyers who do land from a review platform convert at 5-7%, roughly 2-3x organic search and 4-5x paid social, with top-performing inbound lead-to-customer rates of 12-18% reported from review-site sources. lower volume, higher intent: the people still clicking through from G2 or Capterra today are the ones further down the funnel, already comparing vendors within weeks of a decision. a half-star rating improvement measurably lifts click-through from comparison pages and category grids.
the caution worth keeping
review platforms sell paid placements and intent data to the vendors they list, and independent critiques describe parts of the model as pay-to-play. that’s a reason to validate attribution rather than treat a badge as guaranteed lift — track demo requests that cite reviews as a research source, win rate on deals where the buyer referenced the profile, and conversion on pages with versus without review badges, before spending more on placement.
what to check before the next review-program budget
- treat review profiles as an AI-visibility asset, not just a traffic source — the 88% citation concentration means review presence shapes AI answers even when it sends no clicks
- prioritize review volume and freshness on the five platforms AI actually cites — spreading effort across a long tail of review sites misses where the citations concentrate
- measure review-sourced influence with win-rate and demo-request attribution, not site traffic — traffic fell 85-92% while influence did not, so traffic alone now misreads the channel
how this connects to the rest of the stack
this is the same shift covered in the AI-search visibility data elsewhere: ranking number one on Google no longer guarantees being named in an AI answer, and review platforms are one of the main places those answers are sourced from. HubSpot or Salesforce is where “found us via G2” needs a proper source value, since the click often happens long after the review itself influenced the shortlist.
start scaling — if the review-site report still leads with traffic, citation presence and review-referenced win rate are the figures worth adding before the next budget meeting. let’s connect.
Sources: 2026 B2B software review-site and AI search benchmark data (SE Ranking, G2 Buyer Behavior research, Naoma, PageCrawl, Numriq, Stackmatix, The Next Web). Figures are published industry research, not verified results from Amr’s own client accounts.