SQL field guide / Define the denominator
Calculate Monthly Cohort Retention Without Counting Unfinished Months
Count retained users once and separate a completed month with no activity from a month that has not finished.
Lumen / Product analytics Beta
See what happens
after you ship.
See which features people use, follow their sessions, and inspect the events behind a product decision. Open-source product analytics, connected to your SyneHQ workspace.
Included at no additional product charge during beta.
Explore Lumen
Did the feature land?
Follow the activity.
Compare event activity, open a user or session, and inspect the records behind it. This interactive preview uses sample data, so you can explore before connecting your product.
SyneHQ/LumenInteractive demo
132,823 synthetic events · 90 daysOverview
Sample product activity · all times UTC
Events (30d)
+23.3% vs previous 30 days
Unique users (30d)
+1.0% vs previous 30 days
Sessions (30d)
+23.3% vs previous 30 days
Identified users (30d)
+1.0% vs previous 30 days
Events over time
Daily events, last 30 days
View chart data for events over time
| date | events |
|---|---|
| Aug 12 | 1,902 |
| Aug 13 | 1,617 |
| Aug 14 | 1,564 |
| Aug 15 | 1,747 |
| Aug 16 | 1,691 |
| Aug 17 | 1,641 |
| Aug 18 | 1,599 |
| Aug 19 | 1,569 |
| Aug 20 | 1,315 |
| Aug 21 | 1,317 |
| Aug 22 | 1,577 |
| Aug 23 | 1,614 |
| Aug 24 | 1,666 |
| Aug 25 | 1,731 |
| Aug 26 | 1,806 |
| Aug 27 | 1,647 |
| Aug 28 | 1,730 |
| Aug 29 | 2,050 |
| Aug 30 | 2,124 |
| Aug 31 | 2,186 |
| Sep 1 | 2,236 |
| Sep 2 | 2,269 |
| Sep 3 | 2,045 |
| Sep 4 | 2,044 |
| Sep 5 | 2,266 |
| Sep 6 | 2,234 |
| Sep 7 | 2,190 |
| Sep 8 | 2,138 |
| Sep 9 | 2,082 |
| Sep 10 | 1,785 |
Unique users over time
Daily unique people, last 30 days
View chart data for unique users over time
| date | users |
|---|---|
| Aug 12 | 476 |
| Aug 13 | 405 |
| Aug 14 | 391 |
| Aug 15 | 437 |
| Aug 16 | 423 |
| Aug 17 | 411 |
| Aug 18 | 400 |
| Aug 19 | 393 |
| Aug 20 | 329 |
| Aug 21 | 330 |
| Aug 22 | 395 |
| Aug 23 | 404 |
| Aug 24 | 417 |
| Aug 25 | 433 |
| Aug 26 | 452 |
| Aug 27 | 412 |
| Aug 28 | 433 |
| Aug 29 | 513 |
| Aug 30 | 531 |
| Aug 31 | 547 |
| Sep 1 | 559 |
| Sep 2 | 568 |
| Sep 3 | 512 |
| Sep 4 | 511 |
| Sep 5 | 567 |
| Sep 6 | 559 |
| Sep 7 | 548 |
| Sep 8 | 535 |
| Sep 9 | 521 |
| Sep 10 | 447 |
Top events
Most frequent product actions
View chart data for top events
| name | count |
|---|---|
| page.viewed | 16,671 |
| report.viewed | 11,221 |
| filter.applied | 7,160 |
| project.created | 6,026 |
| invite.sent | 4,374 |
| report.shared | 3,821 |
| checkout.completed | 3,404 |
| export.completed | 2,705 |
Top pages
Most viewed product paths
View chart data for top pages
| name | count |
|---|---|
| /workspace | 13,690 |
| /reports/revenue | 7,218 |
| /projects | 5,732 |
| /reports/retention | 5,058 |
| /settings/team | 4,621 |
| /checkout | 4,290 |
| /explore | 3,958 |
| /dashboards | 3,839 |
| /reports/acquisition | 3,540 |
| /onboarding | 3,436 |
Practice the underlying skill
Work through a SQL example.
Activity is a starting point. Learn how a retention query defines eligible cohorts, counts each user once, and handles incomplete months.
Runs locally in DuckDB. Fictional data. No signup required.
From feature activity to product decisions
See what gets used.
Compare feature events and the people behind them. Give your next product review a clear starting point.
Events + users
Follow the session.
Read the sequence around an action. See the recorded steps leading up to a share, a checkout, or an exit.
Session event paths
Inspect the evidence.
Check incoming activity in Live. Use Console to query raw events when the summary raises another question.
Live events + raw queries
Start with one meaningful event
Instrument the action.
Follow what happens next.
Track the action that tells you a feature is being used: a project created, a checkout completed, or a report shared. Add the context you’ll need to understand it later.
Use TypeScript in the browser, Go or Python on the server, or send events over HTTP. Your team decides what to collect.
Browser SDK · TypeScript
npm install @synehq/lumen-jsimport { Lumen } from "@synehq/lumen-js";
const lumen = new Lumen("<ingest-key>", {
endpoint: "https://<your-ingest-host>",
});
lumen.identify("user_123", { plan: "team" });
lumen.track("report.shared", {
report_type: "quarterly",
});Use the ingest key and endpoint from your Lumen setup. Keep administrative credentials on the server.
SyneHQ / lumen
Open source.
Run it on your terms.
Use Lumen in SyneHQ, inspect the code, or host the telemetry server on your own infrastructure.
A server you can inspect. SDKs you can build with.
Server: AGPL-3.0-or-later
Client SDKs: Apache-2.0
What does Lumen cost?
Lumen is included at no additional product charge during beta. This does not establish future pricing or volume terms. The open-source server is also available to self-host under AGPL-3.0-or-later; SDKs use Apache-2.0.
Where is hosted data stored, and for how long?
Each team has a dedicated ClickHouse-backed tenant in us-east-1. Raw event retention is currently one year. Schema and retention behavior may change during beta.
Can I delete my team’s telemetry?
On a tenant deletion request, Lumen purges the data, including backups, within 30 days. Your team controls which events and properties it collects; SyneHQ processes hosted data under its DPA.
Do sessions include video replay?
The session view shows the sequence of recorded events and their properties. It is not a video recording of a user’s screen.
Can I self-host or request EU hosting?
You can self-host the open-source Lumen server. The hosted beta is in us-east-1; contact SyneHQ about EU hosting or Standard Contractual Clauses before relying on it for those requirements.
Ship the feature. See what happens next.
Start with the events behind one important workflow, then follow how people use it.