Build or Buy? A Decision Framework That Actually Helps

We build custom software, so you might expect us to argue for building. We do not, consistently. A good portion of the enquiries we receive describe problems an existing product solves better and cheaper than we could. Telling someone that costs us a project and earns us a reputation, which is the better trade.

Here is the framework we actually use on a first call.

1. Is this process the thing that makes you money?

If the process is how you compete — the thing you do differently from everyone else in your market — custom software can be a genuine advantage. If it is a process every business in the world runs the same way, such as payroll, accounting or email, you are almost certainly better off buying.

The test: if a competitor could copy your process by buying the same software you did, it was never your advantage.

2. Have you actually tried the off-the-shelf option?

"It does not do what we need" is often shorthand for "we looked at it for twenty minutes". Before commissioning a build, run a real trial with real data and real users for a few weeks. You will either confirm the gap is genuine — which makes your custom scope far sharper — or discover that the gap was a configuration setting.

Trials are cheap. Builds are not.

3. Can you live with 80%?

Existing products rarely fit perfectly. The question is whether the misfit costs you more than a build would. Sometimes adapting your process to the software is the right answer, particularly if your process grew by accident rather than design. Sometimes the missing 20% is the entire reason your business exists. Be honest about which one you are looking at.

4. Who maintains it in three years?

Bought software is maintained by the vendor. Custom software is maintained by whoever you pay to maintain it — and if that relationship ends, by whoever you can find. This is the cost most businesses underestimate.

It is also why we choose technology for the client rather than for ourselves. A less fashionable stack you can hire for locally beats an elegant one that only its original author understands.

5. What happens if you do nothing for six months?

If the honest answer is "not much", you have time to trial alternatives properly. If it is "we lose customers" or "we cannot expand", you have a real constraint — and a real constraint makes a much better brief than a vague desire to modernise.

The middle path people miss

Build and buy are not the only options. Very often the right answer is to buy the commodity parts and build only the piece that is genuinely yours — an off-the-shelf accounting package with a custom integration into your operations system, for instance. You get the vendor's maintenance on the boring parts and bespoke behaviour exactly where it matters.

That is usually the cheapest good answer, and it is the one we recommend most often.