Reviews

Notes from people who sat in the room, including the ones who argued with us.

The 2025 Bangkok evening cohort scored the flagship 4.3 out of 5 on the exit survey. That is not a marketing target; it is the number we actually received after two students marked the habit-window module as too slow.

★★★★☆

“Adoption Signals module 03 forced us to stop calling a failed QR payment a ‘trial.’ Our council readout got shorter and harder to ignore.”

Kanchana R. · analytics lead, Bangkok

“The taxonomy waitlist would have saved us a quarter. I joined the flagship anyway; the event map template is now on our wiki with red ink from Nalinee still on page two.”

Wit, Chiang Mai

8.7 / 10 · internal LMS

Would take again: yes. Would recommend to a teammate who only wants SQL tricks: no. Best module: exposure conditions. Weakest: the optional Asoke office hour clashed with my last BTS train.

Anonymous · client in grocery delivery

“I wanted a faster habit window for a monthly insurance add-on. The room would not let me. I still think they were slightly too rigid, and my floor memo is better because of that fight.”

Client in insurance, Hat Yai

Case · retiring in-app ‘Moments’ after the floor missed twice

Workshop participants around laptops

A Bangkok social app shipped a photo-reaction tray to 22% of users behind a flag. After Adoption Signals, the team defined exposure as “tray rendered in an active session,” trial as “first reaction that persisted,” and habit as “reactions on four distinct days in 28.” The floor was 6% habit/exposure, chosen before week-four data existed.

Week four landed at 2.1%. They did not move the floor. A second window after copy changes reached 2.8%. The council used the Stewardship one-pager to retire the tray instead of funding a sticker pack. The reservation from their PM: the course did not help them estimate engineering salvage cost, which they had to add themselves.

Case shared with permission · names withheld

Case · LINE entry points and a false exposure spike

A travel wallet team in the Atelier cohort saw exposure jump after a LINE rich-message campaign. Module 02 critique showed most of those sessions landed on a Thai error state because the English deep link was still live. Re-instrumented exposure (successful render of the Thai transfer sheet) cut the “adoption” story in half and stopped a premature scale-up.

Told by a student who asked to be listed as “Pim, product analytics, Bangkok”