Every few weeks somebody asks us to build something that already exists. Sometimes they are right and we build it. More often the sensible answer is to configure a platform for the ordinary parts and build only the part that is genuinely yours.
The test we use
Ask where your advantage lives. If a process is the reason clients choose you, own it in software you control. If it is plumbing that every organisation in your sector runs the same way, buy it and spend the money elsewhere.
- —Is this process the reason a client picks you, or just the cost of doing business?
- —How often will it have to change once it is live?
- —Who has to be able to change it: your team, or an engineer?
- —What happens to your data if the vendor raises prices or closes?
- —How much of the platform would you have to fight to get what you want?
Where configuration wins
Accounting, payroll, email, standard scheduling, generic storage. Fighting a mature platform to save a licence fee is usually the most expensive decision on the list.
Where bespoke wins
Anything that encodes judgement your competitors do not have: a clinical pathway, a vetting standard, a pricing model, a way of routing work that took you a decade to learn. Those do not fit a dropdown, and forcing them into one quietly flattens the thing you were good at.
Buy the plumbing. Build the judgement.
How we split a system
Most of the systems we ship are both, with a clean line between them, so the bespoke part stays small enough to keep changing cheaply.