We asked the AI assistants your buyers actually use which software they recommend, across 16 go-to-market categories. Here is what they say, who is invisible, and why the answer depends on which AI you ask.
Every tool's position-weighted recommendation frequency across all engines and categories, normalized 0 to 100. A #1 pick counts more than a #5; breadth across categories is rewarded.
| Tool | AI-Visibility | Score | Mentions | #1 pick | Cats | |
|---|---|---|---|---|---|---|
| 1 | HubSpot | 100 | 78 | 36 | 8 | |
| 2 | Ahrefs | 40.2 | 26 | 18 | 2 | |
| 3 | Mailchimp | 33.4 | 32 | 11 | 4 | |
| 4 | Zendesk | 31.3 | 19 | 14 | 2 | |
| 5 | Calendly | 31.1 | 18 | 15 | 1 | |
| 6 | Apollo.io | 30 | 22 | 12 | 2 | |
| 7 | Jasper | 29.1 | 19 | 13 | 3 | |
| 8 | Unbounce | 26 | 18 | 11 | 2 | |
| 9 | SEMrush | 23.2 | 25 | 2 | 2 | |
| 10 | Asana | 22.9 | 16 | 9 | 1 | |
| 11 | Hootsuite | 22.6 | 16 | 9 | 1 | |
| 12 | Buffer | 22 | 17 | 7 | 1 | |
| 13 | ChatGPT | 20.7 | 15 | 9 | 5 | |
| 14 | Outreach | 20.3 | 18 | 7 | 2 | |
| 15 | ActiveCampaign | 19.6 | 24 | 3 | 2 | |
| 16 | SurveyMonkey | 18.7 | 12 | 8 | 1 | |
| 17 | Freshdesk | 18.2 | 17 | 3 | 1 | |
| 18 | Salesforce | 17.9 | 15 | 6 | 4 | |
| 19 | Klaviyo | 17.7 | 24 | 3 | 3 | |
| 20 | Typeform | 17.6 | 16 | 4 | 1 | |
| 21 | Amplitude | 16.8 | 12 | 6 | 1 | |
| 22 | Mixpanel | 16.2 | 14 | 4 | 1 | |
| 23 | Marketo | 16.1 | 16 | 5 | 3 | |
| 24 | ClickUp | 15.4 | 12 | 6 | 2 | |
| 25 | Dialogflow | 14.9 | 10 | 6 | 1 |
Top 25 of 354 measured tools. Linked tools have a full AI-Visibility report card. Complete ranking in the open dataset.
Mostly, no. Below is each engine's top pick in every category. The five engines named the same single best tool in only 0% of categories, disagreeing in 16 of 16. Which AI a buyer happens to ask changes which tool they're told to buy.
| Category | Llama 3.3 via Groqmemory | Cohere Command-Amemory | Gemini 2.5 Flash-Litegrounded | Perplexitygrounded | ChatGPTgrounded | Llama 4 Scoutmemory | GPT-OSSmemory | Qwen 3.6memory | Agree? |
|---|---|---|---|---|---|---|---|---|---|
| AI copywriting | WordLift | Jasper | Jasper | Jasper | Jasper | Jasper | Jasper | Jasper | ≠ |
| CRM | HubSpot | HubSpot | Salesforce | HubSpot | HubSpot | HubSpot | HubSpot | HubSpot | ≠ |
| Email marketing | Mailchimp | Mailchimp | Klaviyo | HubSpot | ActiveCampaign | Mailchimp | Mailchimp | HubSpot | ≠ |
| SEO tools | Ahrefs | Ahrefs | SEMrush | SEMrush | Ahrefs | Ahrefs | Ahrefs | Ahrefs | ≠ |
| Sales engagement | HubSpot | Outreach | Amplemarket | Amplemarket | Outreach | HubSpot | Outreach | Apollo.io | ≠ |
| Product analytics | Google Analytics | Amplitude | Amplitude | Amplitude | Amplitude | Mixpanel | Amplitude | Mixpanel | ≠ |
| Project management | Asana | Trello | Asana | Asana | ClickUp | Asana | ClickUp | ClickUp | ≠ |
| Customer support | Freshdesk | Zendesk | Zendesk | Zendesk | Zendesk | Zendesk | Zendesk | Zendesk | ≠ |
| Social media management | Hootsuite | Hootsuite | Buffer | Buffer | Buffer | Hootsuite | Buffer | Buffer | ≠ |
| Landing page builders | Unbounce | Unbounce | Landingi | Unbounce | Webflow | Unbounce | Unbounce | Webflow | ≠ |
| Lead generation | HubSpot | HubSpot | SyncGTM | Apollo.io | Apollo.io | HubSpot | Apollo.io | Apollo.io | ≠ |
| Marketing automation | Marketo | HubSpot | ActiveCampaign | ActiveCampaign | HubSpot | Marketo | HubSpot | HubSpot | ≠ |
| AI chatbots | Dialogflow | Dialogflow | ChatGPT | ChatGPT | ChatGPT | Dialogflow | ChatGPT | Chatbase | ≠ |
| Scheduling | Calendly | Calendly | Celoxis | Calendly | Calendly | Calendly | Calendly | Calendly | ≠ |
| Survey / forms | SurveyMonkey | Typeform | SurveyMonkey | SurveyNinja | SurveyMonkey | SurveyMonkey | Qualtrics | Typeform | ≠ |
| GEO / AI visibility | Ahrefs | Ahrefs | Ahrefs | Siftly | Profound | Ahrefs | Surfer | None | ≠ |
The split above is not random. It tracks how each engine knows what it knows.
Llama 3.3 via Groq and Cohere Command-A and Llama 4 Scout and GPT-OSS and Qwen 3.6 answer from training data. They reliably name the established, widely-written-about brands (the HubSpots and Mailchimps) and tend to miss anything newer than their training cut-off.
Gemini 2.5 Flash-Lite, Perplexity, ChatGPT read live search results, so they surface the current and trending winners (newer category leaders in email, sales, and AI-visibility tooling) that the memory engines never mention.
AI assistants increasingly answer "what's the best tool for X" before a buyer ever scans a page of links, which makes being named by AI a surface distinct from your Google ranking. The engines we queried are public and reproducible (Llama 3.3 via Groq, Cohere Command-A, Gemini 2.5 Flash-Lite, Perplexity, ChatGPT, Llama 4 Scout, GPT-OSS, Qwen 3.6), each asked the identical question set, with the full raw output in the open dataset below. This audit builds on our earlier AI citation autopsy and the 30-tool GTM index.
When the grounded engines backed their answers with live web search, these are the domains they cited most. This is the map of where to be cited if you want AI to recommend you.
The consensus short list per category, blended across all 8 engines.
Run the AI-Visibility check to see where your product lands when buyers ask AI for a recommendation.
SOURCED: every recommendation is a recorded output of a live AI engine. Engines audited: Llama 3.3 70B via Groq (model memory) [model-memory, 48 queries]; Cohere Command-A (model memory) [model-memory, 48 queries]; Gemini 2.5 Flash-Lite (grounded) [web-grounded, 16 queries]; Perplexity (web search) [web-grounded, 16 queries]; ChatGPT (web, GPT-5) [web-grounded, 16 queries]; Llama 4 Scout (model memory) [model-memory, 48 queries]; GPT-OSS 120B (OpenAI open weights, model memory) [model-memory, 48 queries]; Qwen 3.6 (Alibaba, model memory) [model-memory, 48 queries]. Across 16 B2B/GTM software categories, the two model-memory engines answered all three phrasings per category; the three web-grounded engines (Gemini via API, ChatGPT and Perplexity via browser) answered one query per category, capturing live web sources where available. DERIVED: the AI-Visibility Score (position-weighted recommendation frequency, normalized 0-100), per-category leaderboards, cross-engine agreement, and source-citation tallies. HONESTY FLOOR: this measures what AI engines OUTPUT, not product quality or our opinion. Search-grounded engines reflect the live web; model-memory engines reflect training data and can name tools that do not exist or omit real leaders. Because the grounded engine is sampled less densely than the memory engines, the blended leaderboard leans slightly toward model-memory; the per-engine comparison below separates them. Outputs vary by phrasing, date, and engine; this is a dated capture window, not a ranking endorsement. Browser passes (ChatGPT, Perplexity) are layered in as documented. CAPTURE WINDOW: this dataset is a collection window, not a single snapshot. Five engines were captured 2026-06-19 (Perplexity, ChatGPT, Gemini 2.5 Flash-Lite grounded, Cohere Command-A, Llama 3.3 70B) and three open-weight memory models were added 2026-07-08 (Llama 4 Scout, GPT-OSS 120B, Qwen 3.6); per-engine dates are recorded in metadata.engineStats capturedOn and the interval in captureWindow. Memory models are web-independent, so their answers are date-invariant across the two capture dates; all three grounded/web engines share the single 2026-06-19 date, so there is no temporal confound within the grounded set.
Open data: ai-recommendation-audit-2026.json, free to reuse with attribution to Lucreya. This measures what AI engines OUTPUT, not product quality; a dated snapshot, re-run to reproduce.
Couey, V. W. (2026). The AI Recommendation Audit 2026: Which B2B/GTM Software AI Assistants Recommend [Data set]. Lucreya. https://doi.org/10.5281/zenodo.20767878