In development · Small pharmacies, Bangladesh

The two things that
quietly cost a pharmacy

A small pharmacy in Bangladesh does not lose money on the sale. It loses money on stock that crossed its return window while sitting on a shelf, and on dated debt to four different distributors that nobody is tracking on paper. This is software built around those two problems and almost nothing else.

Bangla: the words the shop already uses, not software wordsWorks offline: the shop's own device, no connection neededTwo problems: near-expiry returns and distributor credit
What it does

The parts that
actually run.

Written in the present tense because each of these has been used. Anything not built yet is in the last paragraph on this page instead.

01

Warns before the return window closes, not after

Stock can go back to the distributor while it is still inside the return window. Once that window passes, the same box is a loss. The app counts the days so nobody has to.

02

Tracks what is owed, to whom, and by when

Different distributors run different terms, from a week to ninety days. On paper that becomes one number in a notebook. Here it stays as separate dated debts.

03

Reads the delivery slip from a photo

Typing a challan in by hand is where shopkeepers give up on software. Photograph it instead, and the rows get matched against the catalogue on the device itself.

04

Speaks the shop's language

বাকি, জমা, দেনা. The vocabulary is the one already written in the paper khata, because a shopkeeper who has never used an app should not have to learn a second vocabulary to use this one.

How it is built

The decisions,
and why.

Every one of these was a fork in the road where the other option was easier. This is the reasoning, so you can disagree with it.

Pharmacy, not general shop software

General shop software in Bangladesh is crowded and there is a free product with millions of registered businesses in it. A free competitor can stay free forever. Pharmacy is a different buyer with a different reason to pay.

The device does the matching, not a server

Matching medicine names runs on the phone, so it works with no connection and the catalogue cannot quietly drift away from what the shop actually stocks.

The shop can add its own medicine

New brands appear every month, so a missing entry must never be a dead end. Anything not in the catalogue can be added in place and used immediately.

Said plainly

This has no customers. It is being built and extended, the catalogue it ships with is incomplete, and the central assumption, that data entry is the reason shops abandon software, has not yet been checked with a real shopkeeper. That check is the next thing that matters, not another feature.

Especially if you run a pharmacy, or supply one.

The rest of what
gets built here.

Three products are live and two of those take money. The service side builds the same kind of thing, scoped to one business.