Dashboards are a dialect. Most product rooms never learn to speak it.
Terminal Syncpoint trains operators who already live inside an app — PMs, analysts, founders — to instrument, read, and argue from behavioural data without borrowing someone else’s slide template.
Heard after the last room
“The event grammar week on Signal Architecture is the only time anyone has made me rename a schema without sounding pious about it. We still argue, but the arguments are shorter.”
“Retention Cartography did not give me a prettier chart. It gave me a window I can defend in a board pack.”
“Identity stitching moved faster than our CRM could follow. Useful, occasionally bruising. I would sit it again with a data engineer in the chair beside me.”
On the timetable
Programmes currently offered
Signal Architecture
Event names, property discipline, and the political work of getting a taxonomy to stick inside a shipping app.
Retention Cartography
Cohort windows, resurrection, and why “D30” is not a personality. Built around subscription and marketplace apps.
Attribution without Theatre
Paid, organic, and the stories teams tell when a click cannot be owned. No last-touch sermons.
How the room is run
We teach from the export, not from the vendor tour.
Bring the stack you already pay for. Mixpanel, Amplitude, BigQuery, a weary warehouse — the curriculum does not rent you a new one. What we insist on is a named event, a named owner, and a sentence you would read aloud to a sceptical CFO.
-
01
Inventory before ambition
Every room begins with the events you already fire, including the embarrassing ones. New tracking is delayed until the old names can be explained.
-
02
A sentence per chart
If a widget cannot be introduced in one spoken sentence, it does not survive the critique. Decoration is treated as a defect.
-
03
Limitation on the table
We keep a written list of what the data cannot see: logged-out traffic, shared devices, store privacy. The list is part of the deliverable.
Figures we will stand behind
Studio notes
Recent writing
Seats
If your app already has users, you are late enough to start.
Write with a short description of the product and the question you cannot answer from the current board.