Off-the-shelf software or a custom application?

The question is usually asked the wrong way round. It is not “standard or custom”, because for almost every process in a small business there is ready-made software, and it is almost always cheaper.

The right question is: what does the twenty per cent that does not fit cost over five years? Because that twenty per cent does not disappear. It becomes a spreadsheet, an extra step, or a person who catches up on something every Friday.

When standard software clearly wins

When your process is fundamentally the same as it is in a thousand other businesses. Accounting, payroll, time tracking, email, ordinary appointment scheduling: all areas where your peculiarities are smaller than they feel.

The advantages are hard to beat. You pay a fraction, you start with something that already exists instead of waiting for it to be built, somebody else handles updates and uptime, and you benefit from improvements built for everyone.

There is also an advantage that rarely gets named: you can stop. Cancelling a monthly licence is a decision; retiring an application you paid to build is a write-off.

When building your own pays

Three situations, and one is enough.

The process is your business. If the specific way you handle something is the reason customers come to you, then standard software is a compromise in the wrong place.

Licence costs run per person and you are growing. At two people that is a footnote. At fifteen it is the largest line on the bill.

The manual work around the software is measurable. If somebody demonstrably spends several hours a week outwitting the tool, you are already paying for a custom build, just spread across salaries instead of an invoice.

The expensive middle route

There is one variant worse than either path: buying standard software and then customising it until it almost looks like a custom build.

You then pay the licence, the customisation, and on top of that the price of having to check after every vendor update whether your customisations still work. That is the situation in which businesses eventually stop applying updates, and from that point the clock is running.

The sums that carry the decision

Take five years and both sides.

For standard software: licence times people times sixty months, plus the time spent on the manual work around it, honestly measured, plus the foreseeable effort of connecting it to your other systems.

For a custom build: the one-off development, plus ongoing care, plus running costs, plus a margin for changes you cannot think of today.

If the two numbers land close together, take the standard software. At roughly equal cost, the option you can walk away from is the better one.

We will tell you that, even though the custom build would be the larger job. Building an application that only narrowly beats an existing solution is a deal that works out for exactly one side, and we would like to still be your point of contact in five years.