From Idea to 1,000 Educators: What Building Biddaan Taught Me About EdTech in Bangladesh
Biddaan started from a simple observation: coaching centers and independent trainers across Bangladesh were running their entire business — live classes, course sales, exams, payments — across five disconnected tools, or worse, out of a notebook and a WhatsApp group. The opportunity wasn’t “build an LMS.” Global LMS platforms already exist. The opportunity was building one for how educators here actually operate.
Offline-first isn’t a nice-to-have, it’s table stakes. A meaningful share of our educators enroll students in person, collect cash payments, and only later reconcile everything digitally. If your platform assumes every transaction starts online, you’ve excluded a large, active, paying part of the market before you’ve even launched. We built offline enrollment and institute management as first-class flows, not an afterthought bolted onto an online-only product.
Multi-tenancy has to be invisible to the tenant. Each educator or institute on Biddaan needs their own branded space, their own student rosters, their own exam banks — without ever feeling like they’re a “tenant” in someone else’s system. Getting the data isolation architecture right early meant we could onboard institute #1 and institute #1,000 on the same codebase without either one feeling like a compromise.
The exam engine is the actual product, even though it looks like a feature. Coaching centers in Bangladesh live and die by exam quality — question banks, audio questions for language learning, negative marking schemes that match local exam boards, and analytics that tell a trainer exactly where a batch of students is falling behind. We under-invested here in our first version and it showed immediately in retention. Once we treated assessments as core infrastructure rather than a checkbox feature, engagement numbers moved in a way nothing else we shipped matched.
Trust compounds slower than you’d like, and faster than you’d expect once it starts. Our first 50 educators came from direct outreach and word of mouth in tight-knit coaching networks. Educator #1,000 came largely from other educators. That’s the EdTech growth loop that actually works in this market — not performance marketing, but one trainer telling another that their students’ exam scores improved.
Building for a specific market’s real operating constraints, instead of porting a generic SaaS playbook, is slower at first and much more durable later. That’s the trade I’d make again.