Quarterly experiment review
Priya Sharma, product manager at Acme Ltd, asked which experiments had concluded with a winner. Memtro listed them with hypothesis, winning variation and key takeaways, each linked to its results.
Eppo + Memtro
Which Eppo experiments are running, and which variation is winning?
Eppo connects Memtro to your experiments and feature flags. It reads experiments by status, tag and date with hypothesis, progress, winner and takeaways, and the diagnostic checks behind a missing result. It also reads flags with environments, variations and allocations, plus metrics and tags. When asked and granted, it can duplicate, refresh or delete an experiment, add a metric collection, create, edit, switch or archive a flag, manage allocations, and create or delete tags. With events on, Eppo tells Memtro when an experiment's status changes, and a job can start on it. Only Eppo admins can create the key.
In Memtro
What you can do with Eppo in Memtro
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right Eppo tool, reads what it needs, and answers with the source cited.
What it does
- 01Lists experiments by status, tag, key or date with hypothesis, variations, progress, winner and key takeaways.
- 02Reads an experiment's diagnostic checks, to say why it has no results or looks wrong.
- 03Lists feature flags with their environments, variations and allocations, and metrics, environments and tags.
- 04When asked and granted, duplicates or deletes an experiment, adds a metric collection, or refreshes results, which runs queries on your data warehouse.
- 05When asked and granted, creates, edits or archives a flag, turns it on or off in an environment, and adds or archives allocations, changing what your applications serve users. It also creates and deletes tags.
- 06With events on, an experiment's status change, a finished results pipeline or a flag change in Eppo can start a job.
Acts when asked
When you ask it to
These tools change something in Eppo. The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
- 01Duplicate an experiment under a new name
- 02Refresh an experiment's results (runs queries on your data warehouse)
- 03Delete an experiment
- 04Add a metric collection to an experiment
- 05Create a feature flag
- 06Rename a feature flag or change its description or tags
- 07Turn a feature flag on or off in an environment
- 08Add a targeting allocation to a feature flag
- 09Archive an allocation of a feature flag
- 10Archive a feature flag
- 11Create a tag
- 12Delete a tag
Events
When it happens
Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Eppo, choosing the same kinds; the tab shows Eppo's own steps. Each event can start a job, with the event as its input. Eppo signs every delivery, and one whose signature does not check out is refused before it is read.
Events and webhooks- 01An experiment's status changed
- 02An experiment's configuration was edited
- 03A metric or metric collection was added to or removed from an experiment
- 04An experiment's results pipeline finished (success or failure)
- 05A flag, its allocations, rules, environments or audiences changed
In practice
Ask it like you would a colleague
Once Eppo is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
Eppo + MemtroWhich experiments are running, and who is winning?
Why does the checkout experiment have no results?
Which feature flags are on in production?
Which metrics are certified, and what is their guardrail cutoff?
Turn the new-pricing flag off in staging.
Pipeline failure follow-up
At BigCo, a job starts when Eppo reports an experiment's results pipeline finished. On a failure, it reads the experiment's diagnostic checks and gives Dan Fletcher, the data lead, the failing check and its error.
Teams
Where it fits
The teams that lean on Eppo most, a question each of them asks, and what they connect beside it.
- Data and analyticsQuestions answered from the warehouse, with the SQL shownAnalyst, Data engineer, Anybody with a question, Head of data“Revenue by segment for Q3 from the warehouse, with the SQL you used”Used alongside Postgres, Stripe, Slack, Google Drive, Snowflake and 3 more.
- MarketingSpend, leads and reputation in one conversationMarketing lead, Performance marketer, Content and SEO, Brand and community“How did last week's LinkedIn and Google campaigns do, and which leads reached the pipeline?”Used alongside Google Ads, Meta Ads, HubSpot, Google Analytics 4, Brevo and 7 more.
- LeadershipOne question across the whole companyFounder, Chief executive, Department head, Board member“What happened with our top ten accounts this week, and is anything at risk?”Used alongside HubSpot, Help Scout, Xero, Stripe, Linear and 6 more.
Better together
Works with
A record in Eppo is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
- Marketing automationEppo with ActiveCampaignContacts, campaigns and automations from ActiveCampaign.
- Marketing automationEppo with BrevoEmail and SMS campaign results from Brevo.
- Marketing automationEppo with Customer.ioCampaign metrics and customer journeys from Customer.io.
- Databases and warehousesEppo with PostgresRead-only SQL over your warehouse or application database, written by the model from the schema.
- Databases and warehousesEppo with MySQL and MariaDBRead-only SQL over your MySQL or MariaDB database, directly or through an SSH tunnel.
- Databases and warehousesEppo with SQL ServerAsk questions of SQL Server and Azure SQL, safely.
- CRM and salesEppo with HubSpotContacts, companies, deals and tickets, with the rest of the company's memory attached.
- CRM and salesEppo with SalesforceAsk your CRM questions in plain English, down to the SOQL.
- CRM and salesEppo with Twenty CRMObjects, fields and records from the open-source CRM, and records it can create and change when you ask.
Also in Analytics
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
- Google Analytics 4
- Plausible Analytics
- Umami
- Google Search Console
- Segment
- App Store Connect
- Google Play Console
- Power BI
- Metabase
- Mixpanel
- Amplitude
- PostHog
- Baremetrics
- ChartMogul
- Fathom Analytics
- Matomo
CallRail
Fullstory- Pendo
WhatConverts- Tableau
- Qualtrics
- Looker
Lightdash- Redash
Microsoft Clarity- Chameleon
Holistics
Mouseflow- Optimizely
- Sigma Computing
- VWO
Setup
How to connect
Connect withREST API key
- 1
In Eppo, open Admin, API Keys and create a REST API key, not an SDK key. Only Eppo admins can create one, and they set its permissions.
- 2
In Memtro, open Connections → Eppo and paste the REST API key.
- 3
An administrator grants the actions by name, such as turning flags on or off, to the teams that may use them. Events are turned on from the connection's Events tab, and the address is then added under Admin, Integrations in Eppo.
- 4
Ask which experiments are running and who is winning.
Also works from
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The Eppo tools are the same everywhere.
FAQ
Questions
Can Memtro change anything in Eppo?+
Only when you ask, and only what it has been granted: duplicate an experiment under a new name; refresh an experiment's results (runs queries on your data warehouse); delete an experiment; add a metric collection to an experiment; create a feature flag; rename a feature flag or change its description or tags; turn a feature flag on or off in an environment; add a targeting allocation to a feature flag; archive an allocation of a feature flag; archive a feature flag; create a tag; delete a tag. Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
Who can use a Eppo connection?+
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Does Memtro store my Eppo data?+
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
How do I connect Eppo?+
REST API key. In Eppo, open Admin, API Keys and create a REST API key, not an SDK key. Only Eppo admins can create one, and they set its permissions.
Can Eppo start a job?+
Yes. Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Eppo, choosing the same kinds; the tab shows Eppo's own steps. A job set to start "When an event arrives" then runs once for each matching event, with the event as its input. The run treats the event as outside data, never as instructions, and anything the run remembers is kept as unverified.
Connect Eppo in a few minutes
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.