Bespoke software vs off-the-shelf: how to choose

We build custom software for a living, so you'd expect us to argue for it. We usually don't. If a £30-a-month product does 90% of what you need, buy it — you cannot beat the economics of software whose development cost is spread across thousands of other businesses.

But there's a specific point at which off-the-shelf quietly becomes the expensive option, and most businesses sail past it without noticing, because the cost shows up as staff time rather than as an invoice. This is how to work out where you are.

Start by pricing what you already do

Before comparing products, price the current process honestly. Take one recurring task — raising invoices, building the rota, reconciling supplier statements, chasing job status — and measure it:

• How many hours per week does it consume, across everyone who touches it?
• How many of those hours are re-typing data that already exists somewhere else?
• What does an error cost when it happens, and how often does it happen?
• How many people know how to do it — and what happens when that person is on holiday?

Six hours a week of re-typing at a £20 fully-loaded hourly cost is about £6,000 a year, every year, forever. That's the number a custom tool has to beat — not the £30 subscription.

When off-the-shelf is clearly right

• The process is standard. Payroll, accounting, email, payments, file storage. These are solved problems with heavy compliance requirements. Never build them.
• You're still figuring the process out. If how you work will change in six months, don't cast it in software yet.
• You need it this week. Custom takes weeks. A subscription takes an afternoon.
• Your workflow is genuinely common. If a product built for your industry fits without workarounds, that's the answer.

When bespoke starts to win

Four signals, in rough order of how strongly they point towards building something:

1. You're paying for the same data three times

A CRM, a quoting tool, a job-tracking sheet and an invoicing package that don't talk to each other. Someone is the integration layer, and that someone is a person on your payroll copying fields between tabs.

2. Your competitive advantage is the workflow itself

If the way you schedule, price or deliver is the reason clients choose you, forcing it into a generic product flattens exactly the thing you're good at. This is the strongest argument for building — and it's why we built ArchiTrak, our own studio-management platform used daily inside UK architecture practices. Nothing off the shelf understood how an architecture practice actually runs a project, so we built the thing that does.

3. The subscription total has quietly overtaken a build

Add up every per-seat licence across every tool that touches one process. Multiply by three years. Compare that to a one-off build you own outright. For a fifteen-person team paying for four overlapping tools, the three-year figure is frequently five figures — and at the end of it you own nothing.

4. You've hit a hard limit you can't design around

The product can't express your pricing rules. It can't produce the report your accountant needs. It can't handle two sites, or two currencies, or approvals in the order you actually approve things. Workarounds are a tax you pay daily.

What bespoke actually costs — and what you get for it

The honest trade is: higher up-front cost, no per-seat licence, and you own the asset. A build that removes six hours a week and costs a few thousand pounds pays for itself inside the first year and then keeps paying. A build that automates something you do twice a year does not — and we'll tell you so on the call rather than take the work.

Things worth insisting on whoever builds it:

• Fixed price, agreed before work starts. Open-ended day rates are how custom software gets its bad reputation.
• You own the code and the data. Non-negotiable. Ask where the data lives and how you export all of it.
• Start with one process, not the whole business. Ship something useful in weeks, then extend.
• It must survive you leaving. A tool only one person understands is a liability dressed as an asset.

A five-minute decision test

Answer these about the process that's annoying you most:

• Does an existing product do 80%+ of it without workarounds? → Buy it.
• Is it standard, regulated or financial (payroll, VAT, accounts)? → Buy it.
• Is a person moving data between two systems every week? → Build a small tool.
• Is the workflow your actual advantage, or a hard limit you can't design around? → Build properly.
• Are you still changing how the process works month to month? → Wait.

Keep reading