Most small businesses that set out to build custom software should have bought something instead, and most of the ones that should have built it are the ones who never considered it. Here is the test that separates them.
Key Takeaways
- Buy when the process is normal for your industry. Build when it is how you win.
- Off the shelf costs a monthly fee. Custom costs a build plus a running cost forever.
- A tool you have to fight every day is a real cost, just a quiet one.
- The worst outcome is paying to customise software that was never going to fit.
The Test
Is this process the reason customers choose you, or is it just something every business like yours has to do?
Payroll, invoicing, appointment reminders and accounting are normal. Someone has built them better than you will, and they have spent a decade on the edge cases you have not thought of. Buy those.
The thing you do differently, the reason a customer picks you over the place down the road, is the thing worth building. It is also usually smaller than you think, and it often sits on top of the software you bought rather than replacing it.
What Each One Really Costs
| Off the shelf | Custom | |
|---|---|---|
| Up front | Setup time, sometimes a fee | The build, $3,000 to $45,000 depending on scope |
| Ongoing | Per user per month, forever, rising | Hosting and maintenance, roughly 10 to 20% of build per year |
| Fit | Eighty percent, on their roadmap | What you asked for, on yours |
| Risk | They change the price or the product | You own it, including the problems |
The Middle Option Most People Miss
You do not have to choose. The common shape for a small business is buying the boring parts and building one thin thing that connects them or fills the single gap that matters. That is usually a few thousand dollars rather than tens of thousands, and it keeps someone else responsible for the payroll rules.
If a salesperson tells you their platform does everything, the honest translation is that it does everything adequately and your one important thing badly.
A Concrete Version
A landscaping company owner wanted a full custom system: scheduling, invoicing, crew tracking, customer portal. The scheduling and invoicing already existed in products he could rent for less than a hundred dollars a month, and they handled tax rules he did not want to own. What did not exist anywhere was the thing his customers actually called about, which was a photo of the finished job sent the same afternoon with the invoice attached. We built that one piece, it connected to the invoicing tool he already paid for, and it came in at the low end of the one-developer band instead of the twenties. The custom part was the part that won him work.
The Honest Counterpoint
Buying has a real ceiling. Subscription pricing rises, per-seat models punish you for growing, and the vendor can change or kill a feature you built your process around. If a tool sits at the centre of your business and the vendor has you locked in, the monthly fee is not the whole cost. That is a genuine argument for building, and it gets stronger the bigger you get. It is rarely the right argument at the start.
Frequently Asked Questions
How do I know if the off-the-shelf tool is good enough?
Use it manually for a month before deciding it is not. Most "it does not do what I need" turns out to be a setting, and the rest becomes a very precise description of the thing worth building.
Is customising an existing platform a good middle ground?
Sometimes, and check the exit before you start. Heavy customisation of someone else's platform can cost as much as a build while leaving you unable to leave.
What about AI tools that build it for me?
Good for the simple end, and see hire a developer vs no code. They break where data has to be right tomorrow.
What if the tool does almost everything I need?
Almost is usually good enough, and the gap is usually worth measuring before you act on it. Write down what the missing twenty percent costs you in hours per week. If the answer is small, live with it. If it is large, that gap is the precise thing worth building, and it is far cheaper to build than a replacement for the whole tool.
The Bottom Line
Buy the normal parts, build the part that is the reason you win, and expect that part to be smaller than you imagined. If you want a written read on which half your idea falls into, the honest read is free. See also build vs buy for startups and what small businesses spend on custom software.
Roberto Espinoza is CEO of Ruzora, which builds custom software for business owners at a fixed price and places pre-vetted senior LATAM engineers with US teams. Get a free honest read on your idea.
