“The recency-window module is the piece I still open before any QBR.”
Flagship program
Cohort Signal Lab
Seven Bangkok evenings. One production segment rebuilt from events to an activation memo. Faculty who have watched those charts fail inside SEA consumer apps.
What you leave with
- A written cohort frame with inclusion, exclusion, and a failure condition
- An event honesty note covering names you will not trust yet
- One activation proposal tied to that segment only
- Faculty critique recording for your team’s internal retro
The lab does not grant a certificate that hiring managers recognize. It grants a segment you can defend. If you need a badge, this is the wrong studio.
Informational price
฿24,900
Lab Bench seat, per person. No checkout on this site. Compare seats or write to enroll.
Remote video seats open only after the Ramintra room is full. Remote participants still submit a live export; synthetic datasets are declined.
Modules
- Evening 01Event honesty Read payloads, not dashboards. Flag events that fire on screen view and on tap without distinction.
- Evening 02Cohort frames versus vanity buckets Write inclusion rules that can empty. If the segment never shrinks, it is not behavioral.
- Evening 03Recency, frequency, and decay Choose windows that match your product’s natural return rhythm, not a template RFM grid.
- Evening 04Identity after ATT Build cuts that survive missing ids, shared devices, and Thailand’s mixed Android park.
- Evening 05Cross-surface leakage Web, app, and LINE entry points that pretend to be one user journey.
- Evening 06Activation from a single segment One product response. Faculty reject stacked campaigns.
- Evening 07Critique night Public defense inside the cohort. Thai-market cases sit beside your own stream.
Instructor
Maliwan Srisuk
Maliwan led growth analytics at a Bangkok super-app before joining Devpacketline as studio faculty. She still reviews partner schemas and teaches the ATT and critique evenings. Her rule: if a PM cannot repeat the segment definition without looking at a slide, the definition is unfinished.
Guest faculty rotate for funnel and taxonomy nights; Maliwan remains the through-line for every Lab Bench intake.
Reviews from this lab
Homework assumed more analyst hours than we had. Critique recordings saved the intake for us, but week four was ugly.
FAQ
Do I need SQL?
You need enough fluency to pull a sanitized event export. The lab is not an SQL class. If your company will not allow any export, wait — we will not accept a fake dataset.
Is there a real limitation I should know?
Yes. Cohort Signal Lab will not make a two-person data team look like a platform org. We do not implement warehouse models, and we do not sit in your sprint after week seven. Teams that cannot assign event owners before evening three usually stall, and we would rather refund per policy than pretend otherwise.
Can my whole product squad join?
Two seats from the same company is the sweet spot. More than four turns critique into a meeting. Use Studio Retainership if you want staggered seats.
What language is the lab in?
English, with Thai used freely in examples and hallway talk. Materials on this site are in English.