Niches · Product analytics & experimentation

Answer a funnel, retention or cohort question

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

0companies, full fit
12companies, fit with a limit
19companies carry the tag
9share of the category's work
5first-hand accounts

What counts as done

Step conversion, retention by period or a cohort size for the same definition as the UI, including cohorts with zero returns, and a cohort created from a query.

Measured, for our own tests: the figures equal the UI's, within 3 calls, and the cohort has the same row count as the query. No docs we read state these figures, so they don't count against a listing here.

How agents ask for it: where do people drop off in onboarding, what's 7-day retention for users who finished onboarding, how long from add-to-cart to checkout, how big is this cohort, make a cohort of power users

Does the job, with a limit 13

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

ListingWhy
Amplitude BB
Amplitude, Inc.
Dashboard REST funnels take ordered or unordered mode, a conversion window (cs), new or active users and cohort segments; retention marks incomplete cohorts. Group-bys cap at 1,000 and cohort-from-query row parity is undocumented. source
Countly E
Countly Ltd
funnels_data returns step conversion with median and p95 time between steps, and retention and cohort tools exist, but they need Enterprise plugins (the Intelligence add-on on Flex) and cohort-from-query row parity is undocumented. source
Fullstory B
Fullstory, Inc.
compute_funnel returns per-step counts and conversion with a conversion window, segment, prior-period compare and dimension breakdown, but funnels are strictly ordered, retention cards can be read but not built, and no cohort-from-query count is offered. source
GrowthBook BB
GrowthBook, Inc.
postFunnelExploration takes steps and a conversion window over a date range, but retention and cohort creation endpoints weren't found, and this is warehouse SQL that the agent's data model must support. source
LaunchDarkly BB
Catamorphic, Co. (dba LaunchDarkly)
query-funnel counts sessions reaching each ordered step with strictOrder and breakdownBy, but returns window totals only, with no time to convert, retention or cohort tool. source
Matomo B
InnoCraft Limited
Funnels and Cohorts APIs expose funnel reports and first-visit cohorts, but Funnels is a premium plugin and the method parameters (conversion window, ordering) render only with JavaScript, so they weren't checked. source
Mixpanel B
Mixpanel, Inc.
Run-Query covers funnels, flows and retention and Create-Cohort makes cohorts, but Run-Query can't filter or break down by cohort, and profile-history (SCD) properties aren't supported. source
Mostly Good Metrics E
Mostly Good Metrics
execute_funnel and execute_retention accept inline definitions in one call, with cohort and return events by day, week or month, but conversion window, ordering options and cohort creation aren't documented. source
OpenPanel B
OpenPanel AB
get_funnel and get_retention_cohort answer funnel and weekly retention questions, but get_funnel lacks time to convert (acc-34, open) and retention drops cohorts with zero returns (acc-35, open). source
Pendo E
Pendo.io, Inc.
queryFunnel computes ordered two- or three-step funnels with segments and exclude-list options, and cohortRetentionCurve builds retention, but funnels stop at three steps and no conversion window or time to convert is listed. source
PostHog B
PostHog Inc. (formerly Hiberly Inc.)
query-funnel, query-retention and their actors tools answer funnel and retention questions, and cohorts can be created, but cohort population from HogQL timed out, failed on extra columns and appended on recalculation (acc-09, open). source
Statsig BB
Amplitude, Inc.
Widget results return funnel steps, conversion over time and time to convert for saved funnel widgets, but ad-hoc funnels, retention and cohort-from-query weren't found in the Console API. source
Userpilot B
Userpilot, Inc.
Funnel completion and retention questions are vendor examples, but no tool schema shows funnel options, retention periods or cohort creation from a query. source

Doesn't do this job 6

Listings in the category that do something else, with the reason.

ListingWhy
Flagsmith B
Bullet Train Ltd
No funnel, retention or cohort tools or endpoints; segments are targeting rules, not analysis cohorts. source
Harness Feature Management & Experimentation BB
Harness Inc.
No funnel, retention or cohort analysis in the FME resources. source
Heap D
Content Square, Inc.
No funnel, retention or cohort API in the developer reference. source
Optimizely Experimentation B
Optimizely North America Inc.
No funnel, retention or cohort analysis in the MCP tools or Feature Experimentation API. source
RudderStack MCP C
RudderStack, Inc.
No funnel, retention or cohort analysis; audiences are warehouse-built segments for activation. source
Woopra D
Woopra, Inc.
The Data Query APIs cover trends, profiles, profile activity and log export; no journey (funnel) or retention report endpoint is listed. source

What stops agents finishing

  • funnel options in the UI (time to convert, ordering) missing from the API or MCP tool
  • cohort query times out in population though it runs in the query API
  • Mixpanel Run-Query cannot filter or break down by cohort

Around it

Before: product-analytics.metric-question. After: company-knowledge.read-research, product-analytics.flag-experiment-change.

Where this comes from

Weight 9 of 100 in Product analytics & experimentation: 5 of 55 account links; vendors (Mixpanel, Amplitude, PostHog) list funnels and retention first among MCP uses, so not lowered further

The same as data: /api/v1/use-cases/product-analytics.funnel-retention.json, or the MCP tool get_use_case with id product-analytics.funnel-retention. Other jobs in this category: /api/v1/niches?category=product-analytics.

Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person. A mark says what the docs show, not what a run measured.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.