We build custom software for a living, so it may surprise you that our first question is often “could you just buy something?” For most routine business needs, the answer is yes. Custom work earns its cost in a narrower set of situations, and it helps to have a clear way of telling which one you are in.
The short version
- Ask whether the task sets your business apart from competitors. If not, buy.1
- Count the whole cost over several years, including being locked in.2
- Buying is usually faster and cheaper to begin with, and custom work can pay back where the gap is costly.
- You can often do both: buy the standard parts, and build only the part that makes you different.
The test that matters most
The software writer Martin Fowler describes a useful split between “utility” and “strategic” software. What separates them, he argues, is whether the underlying business function is a differentiator: if how you do it is a crucial part of what makes you better than the competition, the software supporting it needs to be as good as you can make it.1
His classic example of a utility is payroll. Payroll has to work, but nobody chooses a decorator or a café because of how it runs its payroll.1 Since a utility by definition offers no competitive edge, the obvious thing is to go with a package.1 For a strategic function, a package may limit your ability to stand out.1
| Probably buy | Possibly build |
|---|---|
| Accounting and payroll | The way you price unusual jobs |
| Email and calendars | A customer portal that sets you apart |
| Standard stock control | Your particular way of scheduling work |
| Card payments | Connecting several tools the way only you need |
Those examples are our own, and your answer may be different. A rota can be a utility for one business, but for a care agency with complex rules, it can be the heart of the operation.
The costs people forget
The UK government’s guidance on choosing technology tells its teams to minimise the total cost of ownership, including reducing the chance of getting locked in to long contracts with specific tools and providers.2 Those ideas apply just as well to a small business.
These are the costs we would add up, over three years:
If you buy:
- Monthly fees, multiplied by users, over the whole period
- Price rises, because the first-year rate rarely lasts
- Staff time spent on workarounds, such as re-typing data between systems
- The cost of leaving, and whether you can get your data out
If you build:
- The first build, which is the obvious part
- Hosting, updates and security, every year
- Changes, because your needs will not stand still
- Reliance on the person who built it, unless it is well documented
Custom software is not a one-off purchase. If someone quotes you a price with no mention of what happens afterwards, ask.
Signs you should buy
- Several established products already do about 80% of what you need.
- The process is not what customers choose you for.
- You could change your way of working to fit the tool, and no one would mind.
- You have no appetite for looking after software.
Signs that custom work could pay off
- People spend hours each week copying data between tools.
- You are paying for several products to get one job done.
- Your spreadsheet has become a critical system that one person understands. See what spreadsheet errors can cost.
- Products force you to work in a way that hurts your customers.
- The way you work really is the reason you win work.
A middle path
It is rarely all or nothing. Our favourite approach is often to buy the commodity parts and build a thin layer on top. For example, you might keep your accounting package and card provider, then build one simple screen that brings the information together. The UK government’s guidance makes a related point by encouraging teams to work out what they need to build and what they can buy or get for free,2 and its Technology Code of Practice sets out criteria for designing, building and buying technology.3
Three more ways to reduce risk:
- Start small. Build the most painful part first, and see if it pays back before doing more.
- Own your data. Whichever route you choose, make sure you can get your information out in a usable format.
- Get it in writing. Agree the scope, the price and who owns the result before work begins.
Questions to ask any supplier
Whether buying or building, these questions show up problems early:
- What will this cost me each year, for three years?
- How do I get my data out if I leave?
- What happens if you stop trading, or if the person who built this leaves?
- How do you handle security updates?
- Can I see it working before I commit?
Our honest position
If an existing product fits your business, we will tell you, and you will probably save money. When it does not, that is what our custom software service is for: you talk to the person building it, see a private preview, and agree one price before any work starts. You can also see the kind of thing we build in our live demos.
Sources
-
Martin Fowler, Utility vs Strategic Dichotomy, accessed 7 November 2026. ↩ ↩2 ↩3 ↩4 ↩5
-
Government Digital Service, Choosing technology: an introduction, GOV.UK Service Manual, accessed 7 November 2026. ↩ ↩2 ↩3
-
UK Government, The Technology Code of Practice, accessed 7 November 2026. ↩
Custom software
When off-the-shelf tools nearly fit, but the gap costs you hours every week.
See how it works