Flagship course

Cohort Signal Studio

An eight-week Predictive Retention Analytics studio where you build, critique, and defend a forecast pack for a live or sample product funnel.

8 weeks Live + recordings THB 18,500 informational
Learner studying analytics on a laptop

Modules

  • Week 1 — Cohort hygiene
    Definition windows, exclusion rules, and the quiet ways activation labels lie.
  • Week 2 — Survival thinking
    Curves, censoring, and why averages hide the week you actually lose people.
  • Week 3 — Early-warning features
    Support friction, content skips, payment retries — ranked by lead time.
  • Week 4 — Experiment stress tests
    Feature flags against churn windows; deciding when not to ship.
  • Week 5 — Model humility
    Simple baselines first; documenting uncertainty without theater.
  • Week 6 — Intervention economics
    Attention cost, channel fatigue, and festival calendar collisions in Thailand.
  • Week 7 — Stakeholder critique
    Rehearsed readout with hostile-but-fair questions from mentors.
  • Week 8 — Forecast pack ship
    Final revisions, handoff notes, and alumni circle onboarding.

Learning outcomes

  • Produce a defensible cohort definition and survival view for one product surface.
  • Separate vanity metrics from signals with measurable lead time.
  • Write intervention options with explicit trade-offs and revisit dates.
  • Facilitate a 20-minute critique that surfaces assumptions instead of hiding them.

Informational pricing

Individual seat: THB 18,500. Team discounts for three or more seats are confirmed by invoice. This site does not process payments — contact us to enroll. See refund policy.

FAQ

Do I need a data science background?

Comfort with spreadsheets or basic SQL helps. We teach the judgment layer; we are not a deep learning bootcamp.

What tools do you use?

Worksheets ship in spreadsheet and notebook formats. You may use the warehouse you already have; we do not require a specific vendor.

What is a real limitation of this studio?

We cannot certify that your company will approve instrumentation changes. If your org blocks export access, some exercises will stay on sample datasets — still useful for critique muscle, less useful for your exact funnel.

Are sessions recorded?

Yes. Recordings post within one business day for enrolled learners.