Custom software vs off-the-shelf: how to decide

We sell custom software, so it may surprise you that our default advice is to buy. Accounting, email, payroll and standard CRM are solved problems. Thousands of developers have spent years on those products, and you’d be paying to recreate a worse version. Custom software is a good investment in a narrower set of situations, and this guide will help you tell which side of the line you’re on.

The short version

When off-the-shelf is the right answer

• Your process is standard for your industry, and it isn’t what makes you different.
• A product covers around 80 per cent of what you need, and you could live without the rest or change your process.
• You’re early-stage and your process is still changing. Software built now might be aimed at the wrong target.
• You don’t have someone who can own a software project and make decisions on it.

Test the alternatives properly before deciding that nothing fits. People often rule products out because of a demo they saw, or because of a missing feature which turns out to be available through configuration or an integration.

When custom software earns its cost

These are the benefits of custom business software, in the situations where they actually apply.

• Your process is your advantage. If how you quote, schedule or serve customers is what wins you business, forcing it into a generic tool wears that advantage down.
• You’re paying for workarounds. Staff re-key data between systems, spreadsheets fill the gaps and errors creep in. Those hours are a real, measurable cost.
• Licences are outgrowing the alternative. Per-user pricing becomes painful with large teams, particularly when most of them need only a fraction of the features.
• Nothing fits. Your business model, regulation or data is unusual enough that products don’t cater for it.
• The software is the product. If you’re selling it, you’ll need to own it.

Compare total cost over years, not the first invoice

A fair comparison looks at around five years. For the product, add subscriptions at your expected headcount, likely price rises, implementation and integration work, and the staff time spent on workarounds. For custom, add the build, hosting, and ongoing maintenance and improvements. Agency guides commonly suggest budgeting roughly 15 to 25 per cent of the build cost per year, and our guide on custom software costs looks at this in more detail. Sometimes custom wins easily. Often it doesn’t, and it’s better to know before you start.

Don’t overlook the risks of custom

• You become responsible. Security updates, bugs and upgrades are on you, or on your supplier.
• Supplier dependence. Protect yourself by owning the code and all the accounts, and by choosing mainstream technology that others can pick up.
• Delay and scope creep. Keep the first release small, and make someone responsible for saying no.
• Change management. People may resist a new system, however well it’s built. Involve them early.

The third option: combine them

It’s rarely all or nothing. Many of the best setups use standard products for commodity jobs like finance and email, then connect them together with integration and add a small custom layer for whatever makes the business different. Often this gets you most of the benefit at a fraction of the cost and risk. Low-code tools can also fill smaller gaps, especially for internal tools with few users, although they can bring per-user costs and limits of their own.

A simple way to decide

1. Write down the top five things the software must do.
2. Trial the two or three strongest products against those, using your real data.
3. List what’s missing, and estimate what those gaps cost per year in staff time and errors.
4. Ask whether an integration or small add-on would close the gap.
5. Only then get a custom estimate, and compare five-year totals.

If you go through this and custom still looks right, you’ll have a far better brief for whoever builds it. We’re happy to look at it with you, and we’ll say if we think a product would serve you better.