Bespoke or off-the-shelf software: which fits your business?

Off-the-shelf software is a product built for thousands of businesses, and these days that usually means SaaS: a subscription, per user, per month, hosted by the vendor. Xero, Salesforce, Monday, Shopify. You get a mature product, immediately, with support and updates included, and you accept how it works.

Bespoke software is built for one organisation. You pay for the build, own the result, and it does exactly what you asked, including the awkward step in your process that no product anticipated.

There is also a middle ground worth knowing about: configuring or extending an off-the-shelf platform, or buying most of what you need and building a small custom piece that connects it to the rest. That is frequently the cheapest sensible answer, and it rarely gets proposed by people who only do one of the two.

Work down these in order. The first one that gives you a clear answer is your answer.

1. Is your process standard? Accounting, payroll, email, document storage, general bookkeeping. If yes, buy. Building a worse version of Xero is the most expensive mistake in this whole field.
2. Does a product exist that fits at least 80 percent? If yes, buy it and either accept the remaining 20 percent or build a small integration to cover it. Custom work at the edges is far cheaper than custom work at the centre.
3. Is the missing 20 percent the part that makes you money? If the gap is in the process that differentiates you from competitors, that's the strongest case for building there is.
4. Do the five-year numbers favour building? Multiply your per-seat subscription by your realistic headcount across five years and compare it with a build plus maintenance. Add the cost of manual workarounds. Below about ten users, buying nearly always wins.
5. Can you wait two to three months? If the answer is no because something is on fire, buy something now and revisit the build once the fire is out. Urgency is a bad reason to build and a fine reason to subscribe.

We build software for a living, so take this in the spirit it's meant: there are four situations where we say buy something instead, and we say it before quoting rather than after.

If the job is accounting, payroll or general document management, buy. These are solved problems with mature products, and the regulatory maintenance alone makes building irrational.

If fewer than about ten people will use it and the process is ordinary, buy. The five-year arithmetic doesn't get close, and you'll spend management time on a project instead of on the business.

If nobody can describe the current process without three people disagreeing, fix the process first. Building software around a disputed workflow just makes the disagreement permanent and expensive.

And if the real problem is that people aren't using the tool you already have, more software won't help. That's a training or incentives problem wearing a software costume.

A real example from our own work: PolicyMind exists because policy lifecycle management, with drafting, approval routing, versioning, staff acknowledgement and audit evidence, genuinely wasn't matched by a product at a price the organisations needing it could pay. That's the shape of a good case for building. Its accounting, meanwhile, is off the shelf, like everyone else's.

Most businesses end up here, and it's usually the cheapest sensible outcome: keep off-the-shelf for everything standard, and build only the one process that is genuinely yours, with an integration joining them.

In practice that might mean keeping your accounting package and your email, and building the operational system your staff live in all day, which pushes invoices into the accounting package automatically. You pay for custom development once, where it earns its money, and you never rebuild something a vendor already maintains better than you could.

If that sounds like your situation, the custom software cost guide has the price bands, and the spreadsheet replacement guide covers the most common version of it.