To get your SaaS into the "best tools" lists AI engines cite, start by finding which lists each engine cites for your buyers' prompts. Favor lists that are current and built around a qualifier your product genuinely fits. Pitch editors with facts that make their list better, and treat any paid slot as an ad. There's no single list to win.
We base this on our audit of 25 software-buying prompts across ChatGPT, Perplexity, and Google AI Overviews. The lists each engine cited barely overlapped, and where a list showed a year, it was almost always 2026. So the work is breadth: many current lists, chosen engine by engine.
Key Takeaways
- The lists AI engines cite are a long tail. Across three engines we counted 247 list publishers, and only 4 were cited by all three.
- Cited lists were current: most that showed a year showed 2026.
- General "best X" lists took most list citations on size, integration, and price prompts, except on AI Overviews. Qualifier lists may be a challenger's way in.
- Paid placement needs clear disclosure under FTC guidance. Treat a sponsored slot as an ad.
- Target lists per engine, then pitch at each list's next refresh.
Which "Best Tools" Lists Do AI Engines Actually Cite?
The lists AI engines cite are a long tail, and a different one on each engine. In our audit, lists made up 40.8% of ChatGPT's citations, 81.9% of Perplexity's, and 58.8% of Google AI Overviews'. Across the three engines we counted 247 distinct list publishers, and only 4 were cited by all three.
Our prompts spanned five software categories: CRM, project management, help desk, HR and payroll, and developer monitoring. Each was asked five ways: plain "best X", company size, integration, price band, and "alternatives to" the market leader. ChatGPT ran logged out from the US with search on, Perplexity through its Sonar API, and AI Overviews on US desktop Google. It's one snapshot, one answer per engine per prompt, with rule-based page typing, so read small counts as indicative. The full results are in our guide to GEO for B2B SaaS.
Each engine leaned on different publishers. On Perplexity, forbes.com led with 21 list citations, followed by pcmag.com and techrepublic.com with 14 each and zapier.com with 12. ChatGPT's most-cited list publisher was technologyadvice.com, with 6. On AI Overviews it was salesforce.com, with 9 citations of its own lists.
Below the leaders, the tail is long. ChatGPT's 42 list citations came from 31 publishers, and 25 of those were cited only once. Perplexity's 408 came from 195 publishers, 126 of them cited once. AI Overviews' 87 came from 66 publishers, 57 cited once. Even forbes.com's 21 citations were a small slice of Perplexity's 408. On these numbers, no single publisher dominates any engine's list citations.
Vendors' own lists count too. On Perplexity, 39.9% of named vendors were cited through a list page on their own site. Reddit is a separate channel: AI Overviews cited it in 12 of 21 answers. Our piece on whether ChatGPT uses Reddit to recommend SaaS tools covers how that differs by engine.
Ahrefs' Glen Allsopp analyzed ChatGPT alone, across 750 top-of-funnel prompts about software, products, and agencies. Pooling all three, he reported: "Across all source links, 'best X' blog lists represented 43.8% of all page types." For software alone, the blog-list share was 34%. He also found "35% of cited best lists were published on low authority domains."
Should You Target "Best X for Y" Lists?
You should target "best X for Y" lists, alongside the general ones. On size, integration, and price prompts, 40.0% of ChatGPT's cited lists named the prompt's qualifier in their title or URL. So did 37.6% of Perplexity's and 55.3% of Google AI Overviews'. General "best X" lists carried the rest.
A qualifier is the part of a prompt that narrows it. In our prompts, that meant a team size, "startup", "free" or "affordable", or a tool like Slack, GitHub, or QuickBooks. Qualifiers changed who got named. We compared each answer with the same engine's plain "best X" answer in that category. New vendors made up 38.5% of those named on size prompts, 47.0% on integration prompts, and 38.2% on price prompts. On alternatives prompts it was 53.5%.
That may be an opening for a challenger. When the prompt narrowed, the shortlist changed, and a list written for that narrower need would be a natural page to cite. Our data shows which lists were cited, not why, so treat this as our reading.
Pick qualifiers your product really serves. If you're built for teams of fifty, a list for solo founders is the wrong door, however easy it looks to get onto.
GetApp, for one, runs qualifier lists. On our integration prompts, Perplexity cited GetApp directory pages filtered by a single integration. If your listing leaves out an integration, a filter on it may leave you out too. Our guide to whether G2 and Capterra reviews decide AI recommendations covers that side.
One limit applies to these figures. We matched qualifier words in list titles and URLs, so a list written for startups under a generic title counts as general. The true share of qualifier lists may be higher.
How Current Does a List Need to Be?
In our audit, the lists that got cited were current. Where a cited list showed a year in its title or URL, it was almost always 2026. On ChatGPT, 34 of 42 cited lists, or 81.0%, showed 2026, and none showed 2025 or earlier. That suggests the cited lists were presented as current.
The other engines showed the same pattern with more undated lists. On Perplexity, 56.6% of cited lists showed 2026, 4.7% showed 2025, and 0.7% showed an earlier year. On AI Overviews, 58.6% showed 2026 and 2.3% showed 2025. The rest, 38.0% on Perplexity and 39.1% on AI Overviews, showed no year at all. ChatGPT's cited lists showed the current year most often.
A year in a title is a freshness signal, not an update date. A publisher can change "2025" to "2026" without touching a single entry. So treat the year as a first filter, then check the page itself.
Undated lists aren't a lost cause, though. Close to two in five cited lists on Perplexity and AI Overviews showed no year. Before you rule one out, look for an "updated" line on the page, or current prices in its entries.
Our reading, and it's reasoning rather than measurement: a list that hasn't been refreshed this year is a weaker target. Annual lists get rewritten, and a rewrite is when an editor is most likely to add a product. Time your pitch to it. A list still dated last year is one to watch, since its refresh may be close.
Should You Pay to Get on a List?
Pay for a list placement only if it's clearly disclosed, and even then treat it as an ad, not an endorsement. Some "best X" lists earn money from affiliate fees or sponsored slots. A labeled paid slot is ordinary advertising. An unlabeled one misleads the list's readers.
FTC guidance addresses this directly. Its Q&A on the Endorsement Guides covers sites that earn from affiliate links to a retailer. "You should disclose your relationship to the retailer clearly and conspicuously on your site, so readers can decide how much weight to give your endorsement." It also says, "The closer the disclosure is to your recommendation, the better." That's US guidance, written as questions and answers.
You can often tell how a list makes money. Look for an affiliate disclosure near the top, "sponsored" or "partner" labels, and "visit website" buttons on every entry. Check the publisher's "advertise with us" page, if it has one. Rankings that line up with who advertises on the site are another sign, though harder to prove.
We have no data on how AI engines weigh disclosed or undisclosed paid placements. Our audit typed pages by their titles and URLs, not by who paid for them. So we can't say whether a paid slot helps or hurts your chances of being cited. Anyone who promises that a paid placement will get you cited is guessing.
Our position: an honest listing on a fair list beats a bought top slot. If you do buy a sponsored slot, insist that it's labeled. Ask the publisher how entries are chosen, and whether paid slots are marked, before you pay. In our reading, the vendors all three engines agree on reflect a wider consensus, which one paid slot is unlikely to buy. Our breakdown of how AI engines decide which SaaS vendors to recommend covers the evidence.
How to Pitch Your Way Onto the Lists AI Engines Cite
Start by running your buyers' prompts on each engine and collecting the lists each one cites. Shortlist the current, qualifier-matched ones, then pitch their editors with facts that make the list better. That's listicle outreach aimed at the pages engines already cite for your buyers' questions.
- Run your buyers' prompts on each engine. Use the five styles from our audit: plain "best X", company size, integration, price, and "alternatives to" the leader. Log every list each engine cites. Engines cite different lists, so one that matters on Perplexity may never appear on ChatGPT.
- Shortlist the current, qualifier-fit lists. Deprioritize lists dated last year or earlier unless a refresh looks close. Keep those built for a size, integration, or price your product actually serves. General "best X" lists belong on it too if they're current, since they took most qualifier-prompt list citations on ChatGPT and Perplexity.
- Check how each list makes money. Look for the affiliate and sponsorship signs above, and note which lists are editorial and which are paid. Treat paid ones as advertising, and budget for them that way.
- Give the editor what the list needs. Send a one-line "best for" statement, current pricing, your integrations, and an honest note on who you're not for. That last line makes the rest easier to trust. Make the facts easy to check against your own site.
- Time the pitch to the refresh. An editor updating a list has a reason to look at new entries. Check when last year's version went up, and pitch a few weeks before that date comes around again.
- Publish your own honest "best X for Y" list. Vendors' own list pages were cited on all three engines. Perplexity cited one in all 25 of its answers, and AI Overviews in 14 of 21. Name competitors fairly, or it reads as an ad.
- Re-run the prompts after the refresh. Check whether you made the list, whether it's still cited, and whether the engine now names you. Keep the log, because it becomes your target list for the next round.
Seer Interactive studied 7,683 pages and 47,097 citations across ChatGPT, Gemini, and Perplexity, from March to June 2026. It found that "75% of the pages LLMs cite were updated in the last year." For ChatGPT, it was "73% updated in the last year." The study covered pet retail, vacation rental, energy, and banking, not SaaS. Still, it points the same way as the years on our cited lists, and it's one more reason to time pitches to refreshes.
Frequently Asked Questions
How do I find which "best of" lists AI engines cite?
Run your buyers' prompts on each engine and log every list it cites. The engines differ a lot. ChatGPT, Perplexity, and Google AI Overviews cited lists from 247 publishers across our 25 software-buying prompts. Only 4 were cited by all three, so a list that matters on one engine may not appear on another.
Can I pay to be added to a "best software" list?
Only if the placement is clearly disclosed. FTC guidance says affiliate publishers should disclose the relationship clearly and conspicuously, so readers can decide how much weight to give the endorsement. Treat a labeled sponsored slot as an ad. We have no data on how AI engines weigh paid placements.
Does a list need to be recent to be cited?
Current lists were the norm in our audit. On ChatGPT, 81.0% of cited lists showed 2026 in the title or URL, and none showed an earlier year. According to Seer Interactive's study, 75% of the pages LLMs cite were updated in the last year, though its industries didn't include SaaS.
Should I target "best X for Y" lists?
Yes, alongside general lists. On size, integration, and price prompts, 40.0% of ChatGPT's cited lists named the prompt's qualifier. So did 37.6% of Perplexity's and 55.3% of Google AI Overviews'. Compared with the plain prompt, between 38.2% and 47.0% of the vendors named on those prompts were new.
Do my own "best X" lists get cited?
They can. On Perplexity, 39.9% of named vendors were cited through a list page on their own site. Perplexity cited a vendor's own list in all 25 of its answers. Google AI Overviews cited salesforce.com's own lists 9 times, more than any other list publisher. Keep yours honest: name competitors fairly, and say who your product isn't for.
Conclusion
There's no single "best tools" list to win. Cited lists come from a long tail of publishers, differ by engine, and most dated ones show 2026. Your target is many current, qualifier-fit lists, chosen per engine, with honest facts for their editors. Treat any paid slot as a labeled ad.
Agencies such as Kubnal Bridge specialize in mapping which lists each engine cites for your category. If you'd like that map for your product, we're happy to help you build it.
About the Author
Kubnal Bridge Editorial Team. The Kubnal Bridge editorial team researches and writes about generative engine optimization and AI search visibility across ChatGPT, Perplexity, Google AI Overviews, Gemini, and Claude.
Sources
- Ahrefs (Glen Allsopp), "Do Self-Promotional 'Best' Lists Boost ChatGPT Visibility? Study of 26,283 Source URLs," 2025-12-04, https://ahrefs.com/blog/best-lists-research/
- Federal Trade Commission, "FTC's Endorsement Guides: What People Are Asking," 2023 revision, https://www.ftc.gov/business-guidance/resources/ftcs-endorsement-guides-what-people-are-asking
- Seer Interactive (Sonny Vasquez), "Study: Content Recency's Impact on AI Visibility in 2026," 2026-07-24, https://www.seerinteractive.com/insights/study-content-recencys-impact-on-ai-visibility-in-2026
- Kubnal Bridge, SaaS citation audit: 25 software-buying prompts across ChatGPT, Perplexity, and Google AI Overviews (first-hand data)