Cholo Shikhi Khelar Chholey LMS

When literacy rates are low, language matters. But more than that, the entire interface had to work without text. We replaced text-heavy onboarding with a single tap. Buttons became large, colorful, and used symbols that made sense in the context of a textile factory.

The audio narration was critical. Every action has an accompanying voice instruction in natural Bengali. Not formal or robotic, but the way a family member would explain it to you.

We profiled actual network conditions in factory areas. Most users had between 2G and 3G. Some had nothing. So the app had to work at all levels. Audio content streams and works on the worst connections. Images degrade gracefully. Video only downloads on WiFi and is optional.

The architecture assumes bandwidth will disappear. Users can download lessons locally, work offline, and sync when they reconnect. No one has to know the internet disappeared.

This is one of the most under-appreciated design decisions we made. Phones in Bangladesh aren't individual devices. They're family devices. A mother might have one phone to use while managing household and work. Her child might use the same phone after school. Rather than fighting this reality, we designed with it.

Each profile has its own login, progress, and content recommendations. Switching between profiles takes one tap. No confusion, no lost progress.

We went to the factories. We watched people use their phones. We asked them to use our prototypes and stayed silent when they were confused. That silence told us more than any feedback form could.

The research revealed that garment workers had less than 10 minutes to engage with an app during breaks. That changes everything. You can't have a complex feature set. You have to be ruthless about cutting features that don't serve the core learning goal.

Every feature that made it into the final build had to survive this question: "Can someone with no app experience, on a bad connection, during a break, accomplish this in under 5 minutes?"

Going to Google Play wasn't just about distribution. It was about legitimacy. When a resource appears on the official Play Store, parents treat it differently. It's vetted, it's safe, and it's persistent.

Being on Play also meant reaching families that Save the Children didn't have direct contact with. A mother in a factory across the country could download the app without waiting for a field coordinator to distribute a worksheet.

For the first time, we could track whether learning was actually happening. Not through surveys or reports, but through actual usage data. That feedback loop shaped every iteration.