When your business needs an app — and when it doesn't

The best software projects we see replace a process that already works but doesn't scale: the spreadsheet with 40 tabs, the quoting ritual only Dave can perform, the whiteboard that is, functionally, the company database. The worst ones start with 'we should have an app' and go looking for a reason.

Build when

• A working manual process is drowning in its own volume
• Off-the-shelf tools force workarounds that cost real hours weekly
• Customers keep asking for something a login would solve (status, documents, bookings)
• The process IS your competitive edge — and generic software would flatten it

Don't build when

• A configured off-the-shelf product covers 90% — buy it, integrate it, move on
• The process itself is still changing weekly — stabilise first, then encode
• It's for a hypothetical customer no one has interviewed
• The budget covers version one but not the operating of it — software is a pet, not a statue

Buy for the parts of your business that look like everyone else's. Build for the part that doesn't.

And when you do build: start with the smallest version that changes the Monday morning of the person who'll use it. Portals, dashboards and internal tools earn trust in weeks, then grow. Eighteen-month specifications mostly produce eighteen months of meetings.