GTM AI Index
September 2026 Edition · The permanent record of this edition. The unqualified address always carries the latest edition.
Index › GTM data and infrastructure › September 2026 Edition

Product analytics

Asked as “product analytics platform”, and as “product analytics tool”, on behalf of a mid-market B2B company. 47 first choices recorded across the direct, paraphrase, budget and scale prompts, twelve models each.
Standing
Amplitude
Contested
38% of first choices, but two endorsed leaders sit here: Mixpanel is four points behind on share and eleven points cleaner on negatives.
ShowinginAll 73 categories in the September 2026 Edition →

By buyer segment

The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
Small businessPostHog43%Clear leaderStanding →Mid-marketAmplitude30%ContestedThis pageEnterpriseAmplitude49%Clear leaderStanding →

01The standing

Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all twelve models, for a mid-market B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrant
01Amplitude30%21%62accepted challenger
02Mixpanel28%17%60accepted challenger
03PostHog21%4%56accepted challenger
04Pendo11%11%45accepted challenger
05LogRocket2%10%10accepted challenger
06Google Analytics 42%61%23criticized challenger
Show the two products at 0%, ordered by negative rate
08Heap0%37%41criticized challenger
07Fullstory0%21%14accepted challenger
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its vendor page.
Head to head
Amplitude vs MixpanelAmplitude vs PostHogMixpanel vs PostHog

Recommended versus criticized

Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.

Criticized challengerCriticized default
Negative label rate →
01
02
03
04
05
06
07
08
Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative40%
Key
01Amplitude30%
02Mixpanel28%
03PostHog21%
04Pendo11%
05LogRocket2%
06Google Analytics 42%
07Fullstory0%
08Heap0%

02What they warned about

Five of twelve models held their first choice under the paraphrase. Claude Haiku 4.5, Perplexity Sonar, Grok 4.1 Fast, Mistral Small, Llama 4 Maverick, Qwen 3.7 Flash and Kimi K2 changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
Amplitude
21%
13 of 62 labels negative · 11 of 12 models · 1 hard negative
“Consistently rated as the most expensive option at scale... A poor fit for cost-sensitive teams” Kimi K2, negative prompt
Heap
37%
15 of 41 labels negative · 10 of 12 models · 4 hard negative
“here are the product analytics platforms you should be cautious about or avoid:... **Heap** (by Contentsquare) - **Data inaccuracy**” GLM 4.7 FlashX, negative prompt
Google Analytics 4
61%
14 of 23 labels negative · 9 of 12 models · 8 hard negative
“**Avoid GA4** for EU-based products or if you need product-level insights (it's marketing-focused)” MiniMax M2.5, negative prompt
Mixpanel
17%
10 of 60 labels negative · 9 of 12 models · 3 hard negative
“Highest Caution: Mixpanel ... When to avoid: If handling sensitive user data or EU users” Grok 4.1 Fast, negative prompt

03What they cite

Citations exist only for the models that return a source list: four of the twelve in this edition, and all six flagship models on the expanded tier.

Sites the answers cite

19 of 72 answers in this category came back with a source list, from 4 of 12 models: citations where the model returns them, or the search results it consulted. 6 of those lists are Google grounding redirects that name no site and are left out of the counts. 183 links across 83 sites, every framing counted. Ranked by the number of answers carrying the site or page. The other 8 models searched through a gateway that returned no source list on this edition.

amplitude.com
vendor site · Amplitude10 answers · 14 citations · 3 models
learn.g2.com
10 answers · 11 citations · 2 models
userpilot.com
vendor site · Userpilot8 answers · 8 citations · 2 models
contentsquare.com
vendor site · Contentsquare6 answers · 6 citations · 2 models
visionlabs.com
6 answers · 6 citations · 2 models
cotera.co
5 answers · 5 citations · 2 models
cpoclub.com
5 answers · 5 citations · 2 models
gainsight.com
vendor site · Gainsight5 answers · 5 citations · 2 models
pendo.io
vendor site · Pendo5 answers · 5 citations · 2 models
productschool.com
5 answers · 5 citations · 2 models
checkthat.ai
4 answers · 8 citations · 1 model
mixpanel.com
vendor site · Mixpanel4 answers · 5 citations · 2 models

Pages the answers cite

The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.

learn.g2.com/best-product-analytics-software
10 answers · 2 models
amplitude.com/compare/best-product-analytics-tools
9 answers · 3 models
userpilot.com/blog/product-analytics-tools
7 answers · 2 models
visionlabs.com/blog/best-product-analytics-tools
6 answers · 2 models
contentsquare.com/guides/product-analytics/tools
5 answers · 2 models
cotera.co/articles/product-analytics-platform-comparison
5 answers · 2 models
gainsight.com/essential-guide/product-management-metrics/product-analytics-…
5 answers · 2 models
pendo.io/pendo-blog/top-10-product-analytics-tools
5 answers · 2 models
productschool.com/blog/analytics/product-analytics-tools
5 answers · 2 models
cpoclub.com/tools/product-analytics-tools
4 answers · 2 models

04How they answered

Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ModelDirect“What is the best product analytics platform for a mid-market B2B company?”Paraphrase“Which product analytics tool would you recommend to a mid-sized B2B company?”Comparative“What are the top product analytics platforms and how do they differ?”Budget-constrained“What is the best product analytics platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating a product analytics platform. What should we look at?”Negative“Which product analytics platforms should I avoid or be cautious about?”
Claude Haiku 4.5Mixpanel
Four alternativesAmplitude, Heap, June, Pendo
no first choiceChangedno first choicePostHog
One alternativeGoogle Analytics 4
against: Microsoft Clarityno first choiceagainst: Amplitude, Mixpanel
GPT-5.4 miniAmplitude
Three alternativesHeap, Mixpanel, PostHog
AmplitudeHeld
Three alternativesMixpanel, Pendo, PostHog
Amplitude
Four alternativesHeap, Mixpanel, Pendo, PostHog
against: Google Analytics 4PostHog
Two alternativesGoogle Analytics 4, Mixpanel
against: Amplitudeno first choicenothing named
Gemini 3.5 FlashMixpanel
Three alternativesAmplitude, Pendo, PostHog
MixpanelHeld
Three alternativesAmplitude, Pendo, PostHog
against: Contentsquare, HeapAmplitude
Four alternativesHeap, Mixpanel, Pendo, PostHog
PostHog
Two alternativesAmplitude, Mixpanel
Amplitude, Mixpanel, PostHog
Three alternativesKubit, Mitzu, NetSpring
against: Adobe Analytics, Amplitude, Google Analytics 4, Heap, Looker, Microsoft Power BI, Mixpanel, PostHog, Snowplow, Tableau
Perplexity SonarMixpanel
Four alternativesAmplitude, Heap, Pendo, PostHog
AmplitudeChanged
Two alternativesHeap, Pendo
no first choiceagainst: Google Analytics 4Mixpanel, PostHog
One alternativeGoogle Analytics 4
no first choiceagainst: Adobe Analytics, Amplitude, Contentsquare, Fullstory, Google Analytics 4, Heap, Microsoft Clarity, Mixpanel
Grok 4.1 FastMixpanel
Three alternativesAmplitude, Pendo, PostHog
PendoChanged
Two alternativesAmplitude, Mixpanel
against: Juneno first choicePostHog
Three alternativesAmplitude, Google Analytics 4, Mixpanel
against: Heap, Pendono first choice
Six alternativesAmplitude, Fullstory, Heap, Mixpanel, Pendo, PostHog
against: Amplitude, Heap, Mixpanel, Pendo
Mistral SmallAmplitude Analytics
Three alternativesMixpanel, Pendo, PostHog
MixpanelChanged
Two alternativesAmplitude, PostHog
Amplitude
Five alternativesFullstory, Heap, Mixpanel, Pendo, PostHog
PostHog
Two alternativesAmplitude, Mixpanel
against: Google Analytics 4Amplitude
Three alternativesMixpanel, Pendo, PostHog
against: Amplitude, Google Analytics 4, Heap, Mixpanel, Pendo, VidIQ
DeepSeek V4 FlashAmplitude
Three alternativesMixpanel, Pendo, PostHog
AmplitudeHeld
Three alternativesMixpanel, Pendo, PostHog
against: HeapAmplitude
Seven alternativesContentsquare, Fullstory, Heap, Hotjar, Mixpanel, Pendo, PostHog
against: Google Analytics 4PostHog
Two alternativesAmplitude, Mixpanel
against: Google Analytics 4Amplitude
Three alternativesMixpanel, Pendo, PostHog
against: Heapagainst: Amplitude, Fullstory, Google Analytics 4, Heap, Hotjar, LogRocket, Mixpanel, OpenReplay, Smartlook
Llama 4 MaverickAmplitudeAmplitude, PendoChangedno first choiceCountly, LogRocket, PostHogno first choiceagainst: Amplitude, Google Analytics 4, Heap
Qwen 3.7 FlashAmplitude, Pendo
Two alternativesHeap, Mixpanel
MixpanelChanged
Two alternativesAmplitude, Pendo
Amplitude, Mixpanel
Three alternativesFullstory, Heap, PostHog
Amplitude
Two alternativesMixpanel, PostHog
against: Google Analytics 4no first choiceagainst: Amplitude, Fullstory, Google Analytics 4, Heap, Matomo, Mixpanel, Pendo, PostHog
Kimi K2Amplitude
Four alternativesHeap, Mixpanel, Pendo, PostHog
MixpanelChanged
Three alternativesAmplitude, Pendo, PostHog
Amplitude, PostHog
Five alternativesFullstory, Heap, LogRocket, Mixpanel, Pendo
Mixpanel
One alternativePostHog
against: Amplitudeno first choice
Three alternativesAmplitude, Mixpanel, PostHog
against: Heapagainst: Amplitude, Google Analytics 4, Heap, Mixpanel
GLM 4.7 FlashXMixpanel
Four alternativesAmplitude, Heap, Pendo, PostHog
MixpanelHeld
Four alternativesAmplitude, Heap, Pendo, PostHog
no first choicePostHog
Four alternativesAmplitude, Countly, Heap, Mixpanel
no first choiceagainst: Custify, Dataroid, Flagsmith, Heap, Lucky Orange, Mouseflow
MiniMax M2.5Pendo
Two alternativesAmplitude, Heap
PendoHeld
Two alternativesAmplitude, Mixpanel
Amplitude
Seven alternativesFullstory, Heap, Mixpanel, Pendo, PostHog, Statsig, Userpilot
Google Analytics 4, Statsig
One alternativePostHog
against: Amplitude, MixpanelAmplitude
Two alternativesMixpanel, Pendo
against: Amplitude, Google Analytics 4, Heap, Mixpanel, Pendo
Bold is the first choiceAlternatives are counted; the count opens them.What the answer argued against

05The record

One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency. Full answer text is in the free responses file. Download the record
Seventy-two rows: every prompt, every model, every answer.
PromptModelVersion stringTime (UTC)SearchedSourcesLatency
Direct recommendationClaude Haiku 4.5claude-haiku-4-5-202510012026-09-15 01:52yes1510 s
Direct recommendationGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-15 01:37yes34 s
Direct recommendationGemini 3.5 Flashgemini-3.5-flash2026-09-14 23:53yes2126 s
Direct recommendationPerplexity Sonarsonar2026-09-15 01:28yes205 s
Direct recommendationGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-15 03:12yes08 s
Direct recommendationMistral Smallmistral/mistral-small via mistral2026-09-15 04:22yes010 s
Direct recommendationDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-15 02:59yes022 s
Direct recommendationLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-15 04:15yes02 s
Direct recommendationQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-15 04:10yes054 s
Direct recommendationKimi K2moonshotai/kimi-k2 via novita2026-09-14 23:05yes031 s
Direct recommendationGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-15 00:42yes013 s
Direct recommendationMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-14 23:37yes021 s
ParaphraseClaude Haiku 4.5claude-haiku-4-5-202510012026-09-15 02:08no05 s
ParaphraseGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-14 23:45yes25 s
ParaphraseGemini 3.5 Flashgemini-3.5-flash2026-09-15 02:14yes1426 s
ParaphrasePerplexity Sonarsonar2026-09-15 00:56yes167 s
ParaphraseGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-15 00:07yes08 s
ParaphraseMistral Smallmistral/mistral-small via mistral2026-09-14 23:15yes09 s
ParaphraseDeepSeek V4 Flashdeepseek/deepseek-v4-flash via deepinfra2026-09-14 23:42yes037 s
ParaphraseLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-15 04:04yes02 s
ParaphraseQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-15 03:30yes044 s
ParaphraseKimi K2moonshotai/kimi-k2 via novita2026-09-15 02:05yes022 s
ParaphraseGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-15 02:09yes021 s
ParaphraseMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-15 01:48yes019 s
ComparativeClaude Haiku 4.5claude-haiku-4-5-202510012026-09-14 23:58yes98 s
ComparativeGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-15 02:01yes88 s
ComparativeGemini 3.5 Flashgemini-3.5-flash2026-09-15 01:05yes1024 s
ComparativePerplexity Sonarsonar2026-09-15 00:59yes208 s
ComparativeGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-15 00:41yes07 s
ComparativeMistral Smallmistral/mistral-small via mistral2026-09-15 00:57yes010 s
ComparativeDeepSeek V4 Flashdeepseek/deepseek-v4-flash via deepinfra2026-09-15 00:57yes068 s
ComparativeLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-15 00:38yes02 s
ComparativeQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-15 04:49no034 s
ComparativeKimi K2moonshotai/kimi-k2 via novita2026-09-15 02:24yes030 s
ComparativeGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-15 03:03yes053 s
ComparativeMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-15 02:18yes012 s
Budget constrainedClaude Haiku 4.5claude-haiku-4-5-202510012026-09-15 04:26yes86 s
Budget constrainedGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-15 03:56no02 s
Budget constrainedGemini 3.5 Flashgemini-3.5-flash2026-09-15 02:22yes917 s
Budget constrainedPerplexity Sonarsonar2026-09-14 23:30yes186 s
Budget constrainedGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-14 23:42yes06 s
Budget constrainedMistral Smallmistral/mistral-small via mistral2026-09-14 23:52yes07 s
Budget constrainedDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-15 01:53yes010 s
Budget constrainedLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-14 23:52yes02 s
Budget constrainedQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-15 01:48yes023 s
Budget constrainedKimi K2moonshotai/kimi-k2 via novita2026-09-15 00:27yes033 s
Budget constrainedGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-15 04:02yes061 s
Budget constrainedMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-15 02:35yes011 s
Scale constrainedClaude Haiku 4.5claude-haiku-4-5-202510012026-09-14 23:29no06 s
Scale constrainedGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-15 00:08no011 s
Scale constrainedGemini 3.5 Flashgemini-3.5-flash2026-09-15 01:00yes1423 s
Scale constrainedPerplexity Sonarsonar2026-09-15 02:12yes208 s
Scale constrainedGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-15 00:01yes07 s
Scale constrainedMistral Smallmistral/mistral-small via mistral2026-09-15 02:27yes09 s
Scale constrainedDeepSeek V4 Flashdeepseek/deepseek-v4-flash via deepinfra2026-09-14 23:27yes077 s
Scale constrainedLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-15 00:04yes02 s
Scale constrainedQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-14 23:47yes031 s
Scale constrainedKimi K2moonshotai/kimi-k2 via novita2026-09-14 23:52yes030 s
Scale constrainedGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-15 03:34no060 s
Scale constrainedMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-15 00:25yes013 s
Negative framingClaude Haiku 4.5claude-haiku-4-5-202510012026-09-15 04:09yes2413 s
Negative framingGPT-5.4 minigpt-5.4-mini-2026-03-172026-09-14 23:09no08 s
Negative framingGemini 3.5 Flashgemini-3.5-flash2026-09-15 01:55yes1625 s
Negative framingPerplexity Sonarsonar2026-09-15 02:40yes207 s
Negative framingGrok 4.1 Fastspacexai/grok-4.1-fast-non-reasoning via vertex2026-09-15 04:32yes09 s
Negative framingMistral Smallmistral/mistral-small via mistral2026-09-15 00:51yes011 s
Negative framingDeepSeek V4 Flashdeepseek/deepseek-v4-flash via fireworks2026-09-15 04:43yes023 s
Negative framingLlama 4 Maverickmeta/llama-4-maverick via bedrock2026-09-15 03:22yes02 s
Negative framingQwen 3.7 Flashalibaba/qwen3.7-flash via alibaba2026-09-15 01:21yes038 s
Negative framingKimi K2moonshotai/kimi-k2 via novita2026-09-15 00:44yes062 s
Negative framingGLM 4.7 FlashXzai/glm-4.7-flashx via zai2026-09-15 03:36yes044 s
Negative framingMiniMax M2.5minimax/minimax-m2.5 via minimax2026-09-15 04:35yes023 s

Noise floor in this category

Flips between the edition run and its calibration repeat. Six prompts per model is a small sample; the index-wide floor is the number to trust.

Claude Haiku 4.5
3 of 4 flipped
GPT-5.4 mini
2 of 4 flipped
Gemini 3.5 Flash
2 of 5 flipped
Perplexity Sonar
1 of 3 flipped
Grok 4.1 Fast
5 of 6 flipped
Mistral Small
4 of 6 flipped
DeepSeek V4 Flash
3 of 6 flipped
Llama 4 Maverick
1 of 3 flipped
Qwen 3.7 Flash
5 of 5 flipped
Kimi K2
6 of 6 flipped
GLM 4.7 FlashX
1 of 4 flipped
MiniMax M2.5
4 of 6 flipped

Normalization in this category

Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.

Category-scoped readings
None. Every name in this category resolved on its own.
Unresolved, counted raw
Dataroid
Lucky Orange
Mouseflow
PostHog EU/self-hosted
Discontinued, still offered
No shut-down product was recommended here.

Also in gtm data and infrastructure

The other categories of this function, with the leader in each.
← IdentityTag management →