# Product analytics: what fourteen AI models recommend, and why, November 2026

Gane Index, November 2026 Edition. Page: https://www.gane.ai/index/gtm/gtm-data/product-analytics/guide

Fourteen AI models were asked for product analytics platform six ways each, on behalf of a small, a mid-market and an enterprise B2B company: 252 answers, in which a judge labeled 64 products. Three of them carry at least 10 labels and a first choice, and are ranked. This guide walks the top three in the order the mid-market standing puts them, with what the models said for and against each, in their words. Amplitude leads at 46% of first choices, a clear leader.

## At a glance

| # | Product | Share | Negative rate | Labels |
|---|---|---|---|---|
| 1 | Amplitude | 46% | 13% | 71 |
| 2 | PostHog | 33% | 6% | 71 |
| 3 | Mixpanel | 22% | 16% | 75 |

## 1. Amplitude

Amplitude is #1 of 3 at 46% of first choices, 71 labels, negative rate 13%. By buyer: small business #3 of 4 at 5%; mid-market #1 of 3 at 46%; enterprise #1 of 5 at 61%.

- For: "Amplitude: Strong all-rounder for behavioral analytics, AI insights, experimentation, and enterprise governance. Best if you need depth + scalability." (Grok 4.20, scale prompt)
- For: "The Gold Standard for Deep Insights: Amplitude ... widely considered the leader in the space for large-scale B2B SaaS companies." (Qwen 3.7 Flash, paraphrase prompt)
- Against: "Why to avoid: Amplitude receives a high share of negative labels (31% in one analysis)" (Mistral Small, negative prompt)

## 2. PostHog

PostHog is #2 of 3 at 33% of first choices, 71 labels, negative rate 6%. By buyer: small business #1 of 4 at 62%; mid-market #2 of 3 at 33%; enterprise #5 of 5 at 4%.

- For: "For a limited budget, PostHog (open-source, generous free tier, all-in-one features) is often the best starting point." (GLM 4.7 FlashX, budget prompt)
- For: "For most lean startups today, PostHog is widely recommended as the best balance of price, power, and flexibility." (Qwen 3.7 Flash, budget prompt)
- Against: "PostHog's interface assumes more technical comfort, with the HogQL query editor being powerful but requiring SQL knowledge" (Claude Haiku 4.5, negative prompt)
- Against: "its broad all-in-one toolkit can create operational complexity. Be cautious if you lack engineering ownership" (Perplexity Sonar, negative prompt)

## 3. Mixpanel

Mixpanel is #3 of 3 at 22% of first choices, 75 labels, negative rate 16%. By buyer: small business #2 of 4 at 25%; mid-market #3 of 3 at 22%; enterprise #2 of 5 at 22%.

- Against: "Mixpanel has received significant attention following a data breach in November 2025 that exposed profile information for some API users" (MiniMax M2.5, negative prompt)

## Questions the record answers

**Which product analytics platform do AI models name first most often?** Amplitude, in 46% of first choices for a mid-market B2B company in the November 2026 Edition, from 71 labels. The index calls that a clear leader.

**Does the answer change with the size of the company?** Small business: PostHog at 62%. Mid-market: Amplitude at 46%. Enterprise: Amplitude at 61%. Each standing is computed within its segment and never pooled.

**Which framing changes the answer?** The direct, paraphrase, budget and scale framings count toward share. The product named first differs by framing: direct Amplitude; paraphrase Amplitude; comparative Amplitude; budget-constrained PostHog; scale-constrained Amplitude; negative Fathom. The table above has the counts.

**How stable is one answer?** The repeat measurement is published with the next edition.

**Where do the answers come from?** Fourteen of the fourteen models return sources. Their 70 answers here cite 716 pages; the sites cited most are amplitude.com, mixpanel.com, linkedin.com. The category page lists them all.

**Can a vendor pay to be on this list?** No. A product is on the record because a model named it. A vendor can claim its page, propose corrections to the vendor table and be told when its standing moves; it cannot change a label, a share or a rank, and the publisher's conflicts are disclosed on the method page.

Published under CC BY 4.0. The output is the models' output; nothing here is a recommendation by the index.