About the studio
We exist so upgrade conversations stop living in someone’s head.
Backup Nodehub is an independent teaching practice in Bangkok. We work with product, data, and growth people who already ship subscription apps and still cannot explain why a plan change happens.
Mission
Our mission is narrow: raise the quality of app analytics around upgrades and plan conversion. We are not a general data bootcamp. We do not teach acquisition bidding, brand tracking, or crash reporting. Those crafts matter; they are simply not ours.
Thailand is not a side note. The studio is physically in Bangphlat, and the examples we use — PromptPay onboarding, regional App Store pricing, VAT on digital services — come from work we see here and across neighbouring markets.
Pedagogy
Each course is built as a sequence of readings, not a tool tour. Students bring a live product (or a carefully reconstructed one) and leave with a written diagnostic. Lectures exist to name objects; the homework is to apply those names to a real surface.
We prefer small rooms. Lab Cycle seats are capped so instructors can actually read the event dictionaries students submit. If a cohort is full, we wait rather than dilute the desk time.
Community
Alumni stay on a quiet mailing list used for studio hours, not promotions. Twice a year we host an evening in Bangkok where teams present a single chart they no longer trust — and what they replaced it with. Attendance is optional. The tone is closer to a reading group than a meetup with lanyards.
Desk hours
Lab alumni can book a 40-minute desk hour in the month after their cycle ends. We review one artefact, not a whole roadmap.
Shared lexicon
Graduates keep the paywall taxonomy booklet. Teams that hire two alumni often report they argue less about what “upgrade” means.
Regional peers
Seats regularly mix Bangkok, Chiang Mai, Singapore, and Manila product groups. The contrast is part of the teaching.
Student journey
Most people arrive after a board slide that claimed an upgrade rate nobody in the room could defend. The first conversation is diagnostic: do you need a two-day taxonomy workshop, a four-week trial course, or the full Plan Conversion Analytics Lab?
During the lab, weeks one and two are deliberately slow. We refuse to “instrument first.” Students map every plan a user can buy, every paywall variant currently live, and every billing exception the finance team already knows about. Only then do we design events.
The closing artefact is a ten-page readout. It must include a limitation section — what the analysis cannot claim. We have failed students who delivered a glossy dashboard and an empty limitation. That is by design.
After the cycle, some teams continue into a Residency. Others simply keep the lexicon and write to us when a new plan launches. Both outcomes count as the work succeeding.