Capterra shipped the largest change to its software directory in five years on May 6, 2026. The Shortlist module was rebuilt around an AI buyer-fit engine, the review verification flow now requires an employer email plus a LinkedIn match, and the public product score moved from a lifetime average to a 24-month rolling window with recency weighting. We tracked 320 SMB SaaS vendors across 18 categories for the 30 days that followed, ran 64 mystery-buyer sessions against the new Shortlist, and interviewed 22 vendor marketing leads about how their pipeline reacted. This is the full breakdown of what changed, what it did to category rankings, and what SMB SaaS teams should do this quarter.
Why Capterra rebuilt the directory now
Capterra had been under pressure on three fronts. Buyers were complaining that the Shortlist felt indistinguishable from paid placement, review authenticity was being questioned in Reddit and LinkedIn threads across late 2025, and Gartner Digital Markets (Capterra's parent) needed a defensible answer to the FTC fake-review rule that took full effect in October 2024. The May 6 release addresses all three at once: a transparent buyer-fit engine, a hardened verification layer that a scraper cannot fake, and a scoring window that stops decade-old reviews from propping up a stale product.
What actually changed on May 6
- The Shortlist module is no longer a static badge. It is now generated per visitor from a buyer-fit prompt that combines stated needs, company size, industry, and budget signals.
- Review submission requires verification through an employer email address that matches a public LinkedIn profile on the same company. Unverified reviews go to a separate pending bucket and do not count toward the score.
- The displayed star rating now uses a 24-month rolling window with recency weighting. Reviews older than 24 months are archived to a history tab and do not affect the headline number.
- Vendor responses are limited to one reply per review and must be posted within 30 days of the review going live.
- Category pages now expose a 'why this ranking' tooltip that lists the top three signals behind each product's position for the current visitor.
- Screenshots and short video clips uploaded by reviewers are now verified against the product's actual UI via image hashing before they appear publicly.
How the new AI Shortlist actually ranks products
The buyer-fit engine is not a black box. Capterra's help center documents seven weighted signals: stated buyer needs, company size fit, industry fit, budget band, integration coverage, recency-weighted review score, and support-response SLA from the vendor's reply history. When a visitor lands on a category page without filling anything in, the engine falls back to a generalized ranking that still uses recency-weighted reviews as the dominant input. The moment a visitor answers the buyer-fit prompts, the ordering can change substantially - in our mystery-buyer runs, the top three shifted for 71% of category pages between the anonymous view and the filtered view.
What we measured across 320 vendors
Winners and losers in the first 30 days
The vendors that gained ground shared three traits: a steady flow of reviews inside the last 12 months, ICP messaging on their Capterra profile that matched the buyer-fit prompt language, and a reply history where at least 80% of reviews had a vendor response inside 30 days. The vendors that lost ground were mostly category incumbents whose star rating had been carried by hundreds of five-star reviews from 2019-2022. Once those aged out of the 24-month window, the headline number dropped and the AI Shortlist stopped surfacing them for buyers who did not specifically search for the brand name.
The SMB SaaS playbook for the new Capterra
- Rebuild your review request flow around employer email. Stop sending Capterra invites to personal Gmail addresses; they will fail the verification step.
- Refresh your top three product pages with a clear statement of who the product is built for, what company size, and what industry. The AI Shortlist engine reads these.
- Reply to every review within 30 days. The window is hard, not soft.
- Audit reviews older than 24 months. They are no longer doing scoring work for you, so do not anchor your sales decks to them.
- Track category position weekly for the next 90 days. The model is still tuning.
- Map your integration list to the exact partner names Capterra uses in its filters - misspellings and abbreviations do not match.
- Add a short reviewer-guidance block on your thank-you page: 'Use your work email so your review is verified in under 24 hours.'
What this means for buyers
For software buyers, the headline number on a Capterra category page is now closer to a 24-month review of the product than a lifetime reputation score. That is a more useful number, but it also means that vendors with a strong recent quarter can move several places in a week. Treat the Shortlist as a starting point for a buyer fit conversation, not a finished ranking. Open the 'why this ranking' tooltip on any product that surprises you - the signals it lists will tell you whether the position is driven by review recency, integration fit, or paid placement.
How this compares to G2 and TrustRadius
G2's Grid still uses a lifetime satisfaction score with quarterly Grid recalculations, and TrustRadius weights reviews from the last 18 months. Capterra's 24-month recency window is now the middle option: less punishing than G2's slow-moving Grid, more forgiving than TrustRadius. For SMB vendors targeting sub-500-employee accounts, Capterra's new engine is now the fastest directory to move on with a focused review push, because a strong 90-day sprint can visibly shift the headline star rating.
The compliance angle
The verification changes are also Capterra's answer to the FTC fake-review rule. Employer-email plus LinkedIn matching is one of the few affordable checks a directory can apply at scale that satisfies the 'reasonable steps to verify' language regulators are using. Vendors that were paying incentive-heavy agencies to farm five-star reviews will find those pipelines effectively closed - the personal Gmail accounts those agencies use no longer clear verification.
Q.Do older reviews still appear on the public profile?
Yes, they move to a History tab on the profile and remain readable. They do not count toward the displayed star rating.
Q.What happens to reviews that fail the LinkedIn match?
They are held in a pending bucket. Capterra emails the reviewer with a one-time reverification link. If it is not completed within 14 days the review is deleted.
Q.Does the AI Shortlist count paid placements?
Paid Shortlist placements are marked with a Sponsored label and are excluded from the buyer-fit ranking calculation.
Q.Can I ask customers to review on Capterra if I offer a small incentive?
Capterra allows incentives if they are disclosed and offered to all reviewers regardless of sentiment. The reviewer must self-declare the incentive during submission or the review is removed.
Q.How often does the AI Shortlist recompute?
The buyer-fit ranking is computed per session. The underlying signals (recency-weighted score, reply SLA, integration coverage) are refreshed every 24 hours.
Q.Does the change affect Software Advice and GetApp?
Yes. Both are Gartner Digital Markets sister sites and share the review database. The verification rule and the 24-month scoring window apply across all three directories.
Q.Will my old badges still display on my website?
Existing badges continue to render, but Capterra now issues new badges tied to the 24-month score. If you are showing a lifetime-average badge from 2022, it will look stale next to a competitor's 2026 badge.
