ImportBrain

ImportBrain started as the spreadsheet I kept for GearDockGH. It worked until a single batch arrived in two shipments with three separate clearing charges, and I could no longer say what one power bank had cost me. The first version of the software did exactly one thing: divide a batch into per-unit cost correctly.

Making the batch the central object, rather than the product, was the decision everything else followed from. Inventory systems built elsewhere assume a purchase price per line. Importing does not work that way: freight is quoted per carton, duty is assessed per consignment, and the exchange rate that matters is the one paid on the day the supplier was settled, not the one showing this morning.

If the number on the screen is not the number that was actually paid, the software is worse than the exercise book.

So rates are stored against the transaction. A live rate appears on the dashboard for context and is never used in a historical calculation. It is a small rule that removes a whole class of quietly wrong reporting.

Multi-tenancy went in early even though there was one tenant. Retrofitting isolation onto a working single-user system is the kind of migration that eats a month, and the intended customer — an importer with their own supplier list and their own margins — will not share a database with a competitor on faith.

The dashboard shows zeroes for a new account, and I decided not to seed demo data. An importer who logs in to fabricated numbers learns nothing about whether the tool fits their business. The empty state is honest, and the first real batch fills it in an afternoon.

It is labelled early access because it is. Suppliers, shipments and landed cost are solid; reporting is thin, and there is no offline mode yet for the days the network in the north gives up. Those gaps are stated to anyone who signs up, before they sign up.