How much does custom software cost? An honest guide for small businesses

It's the question every business owner wants answered before they'll take the conversation seriously, and it's the question the industry is worst at answering honestly.

The evasive answer is "it depends," which is technically true and practically useless. The dishonest answer is a number designed to get you to the next meeting. The honest answer requires a little more work, but it's the one that's actually useful.

Here is what I've learned after building custom software for small and medium businesses.

What drives the cost

Software is priced by the time it takes to build, and time is driven by complexity. The variables that matter most:

How much needs to be built versus configured. A simple form that captures enquiries and emails them to you is a few hours of work. A customer portal with login, history, and document management is weeks. Knowing roughly what you're building is the first step to knowing what it costs.

Whether it needs to connect to other systems. Integrations add time. If the new software needs to talk to your accounting system, your CRM, or your booking platform, each one is a piece of work in its own right, with its own quirks and its own testing overhead.

Who uses it and how. Software used by trained staff is simpler to build than software used by customers you can't support directly. Customer-facing software needs to be more intuitive, better tested, and more forgiving of unexpected inputs.

Whether it needs to scale. If ten people will ever use it, the architecture is straightforward. If ten thousand people might use it simultaneously, it's a different problem with a different price tag. Most small businesses are firmly in the first category.

Rough ranges

Rather than dodge this with vague gestures, here are honest starting points:

A simple internal tool or workflow automation: £1,500 to £5,000. Something that saves your team time, captures data more reliably, or replaces a process currently living in a spreadsheet.

A customer-facing application: £5,000 to £20,000. A booking system, a member portal, a self-service tool with authentication and history.

A more complex platform with integrations and ongoing operation: £20,000 and up. This is where you're building something genuinely bespoke, probably with multiple moving parts and a requirement for ongoing maintenance as the business changes.

These ranges assume one experienced developer doing the work. An agency will quote more, because their overheads require it. That's not necessarily a reason to avoid an agency, but it's a reason to understand what you're buying.

Why day rates are misleading

A developer who quotes you a day rate is giving you half the information you need. The other half is how many days. Without a defined scope, a day rate tells you nothing about what you're going to spend. What you want is a fixed price against a defined piece of work, so you know the number before you commit to anything.

Fixed pricing requires a proper discovery phase. Someone has to understand the problem well enough to scope the solution. That might be a paid session or a detailed conversation that leads to a written proposal. Either way, it's the step you shouldn't skip.

Ongoing costs

Software doesn't stop costing money when it's built. The things that continue:

Hosting and infrastructure: usually small. A well-built application for a small business costs tens of pounds a month to run, not hundreds.

Maintenance and support: only necessary when the system genuinely needs it. If someone wants to sell you a retainer for a system that runs fine on its own, that's worth questioning. A good system should be able to sit quietly for months without needing attention.

Changes as your business changes: unavoidable, but controllable. Well-written code means changes are predictable to scope and cost. Owning your code means you're not paying whatever rate the original vendor decides to charge when you need something updated.

When off-the-shelf is cheaper

Honestly: often. If your problem is common and well-served by the market, there's probably an existing tool that handles it for a monthly fee that's a fraction of what a bespoke build would cost. Custom software makes sense when the gap between what the market offers and what your business needs is costing you more than the build would. Not before that point.

I tell people this in discovery conversations. If an off-the-shelf tool is the right answer, I'll say so and point you toward it. Building something you don't need isn't good for either of us.

If you want a straight answer about what your specific problem might cost to solve, starting with a conversation is the fastest way to get one. No quote before a proper scope, but no endless process either: just a clear description of what I'd build and what it would cost before you commit to anything.

What's been the most surprising thing about a software quote you've received? I'd like to hear about it in the comments.