Building useful software, one product at a time

Most of the operational software a mid-sized Indian company runs on is a spreadsheet with someone's name on it. Employee records in one file, attendance in another, leave approved over chat and reconstructed at the end of the month, payroll assembled by hand under time pressure. None of that work is difficult. All of it is expensive, because it consumes the week of the people who could be doing something better.

We started with HR because it was the clearest version of that problem, and because the cost of getting it wrong is immediate and personal: someone is paid late, or paid incorrectly, and finds out on payday.

Workforce is what came out of that. It handles employee records, attendance, leave, payroll and employee self-service as one connected system rather than five disconnected ones. Finpilot, OPM Billing, OPM CRM, Airform and OPM Projects followed the same way, from the same kind of problem: a month end reconstructed by hand, invoices chased by phone, a pipeline tracked in a spreadsheet, answers collected on paper, and project work nobody could see the state of. All six are in production and supported.

What goes into each one is decided mostly by customers telling us what broke this month. That loop is deliberately narrow. We would rather six products be genuinely dependable than a dozen be approximately finished, which is why a product reaches this site when you can buy it and not before. What we intend to build next is set out on What’s next, plainly marked as not available yet.