Ready software or custom build: how to decide
A practical way to choose between a ready-made product, a small customisation, and building something from scratch.
· Nexbiz
Most small businesses ask the question the wrong way round. They start with “what should we build?” when the first question is “does something already exist that gets us 80% of the way there?”
Custom software is the most expensive and slowest option, and often the least reliable in year two when the person who built it has moved on. It is the right answer sometimes — but it should be what you land on after ruling out the alternatives, not where you start.
Here is the order we work through with people.
1. Is there a ready product for this?
For common back-office work — accounts, billing, stock, admissions, appointments, leads, maintenance — ready products exist and have been refined over hundreds of businesses. They cost a fraction of a build, you can see them working before you pay, and someone else maintains them.
The trade-off is that you adapt slightly to the software instead of the other way round. For 90% of businesses that is a fine deal, because your competitive edge is not in how you format an invoice.
Choose ready when: the process is standard, you want to start this month, and budget matters.
2. Can a customisation close the gap?
Sometimes a ready product is right except for two or three things — an extra field, a specific report, a different approval flow, an integration with a tool you already use. That is a customisation, not a build. It is measured in days, not months, and you keep all the benefits of a maintained product underneath.
Choose customisation when: a ready product fits the shape of your work but misses specific, nameable details.
3. Is the process genuinely unusual?
If your operation really does work in a way no product supports — because the business model is unusual, or the workflow is the thing customers pay you for — then a custom build makes sense. Go in with eyes open: fixed written scope, staged delivery, and source code plus documentation handed to you at the end so you are never dependent on one developer.
Choose custom when: the process is core to your business and no product comes close.
A quick test
Write down the five things the software absolutely must do. If a ready product does four of them and the fifth is a report or a field, you are looking at a customisation. If it fails on two or three of the five because of how the work fundamentally flows, that is when a build earns its cost.
Start narrow, prove it works, and expand from there — whichever route you take.