Custom software or off-the-shelf? How to decide

Every growing company eventually hits the point where its software doesn’t quite fit. The CRM can’t handle how you price, the operations team runs a critical process in a spreadsheet, or customers want a portal no product provides. The question is whether to adapt to an existing product, or to build something of your own.

When should you buy off-the-shelf?

Buy when the need is common and well served: accounting, payroll, email, standard CRM, help desk, HR. These products are mature, secure and improve every year at no extra effort from you. Adapting your process slightly to fit a good product is usually cheaper than building.

When does custom software pay off?

• It is how you compete. A pricing engine, a customer experience or an operations model that differs from your competitors’ is worth owning.
• No product fits without heavy workarounds. If staff spend hours each week bridging gaps between tools, the gap is costing money.
• You need to connect systems in a specific way. A custom layer that joins your CRM, finance and operations data can be the highest-value software you own.
• Licence costs scale badly. Per-user pricing for many occasional users can exceed the cost of a focused custom app.
• You need control over data location, security or the roadmap.

What about a hybrid?

Often the best answer. Keep standard products for standard jobs, and build around them: integrations between systems, a customer or partner portal, a mobile app for field staff, or automation for the steps the products don’t cover. You get a reliable core without bending your business to fit it. Our guide to business process automation shows where that layer pays off.

How do you compare the costs?

Look at three to five years, not the first invoice:

• Off-the-shelf: licences at your expected user numbers, configuration, integrations, training, and the hidden cost of workarounds.
• Custom: design and build, hosting, maintenance and updates (often a meaningful share of the build cost every year), security, and the team or partner to support it.

Add the cost of doing nothing: the hours lost each week to the current workaround.

How do you run a custom software project well?

1. Define the problem, not the features. “Field engineers re-type job notes into three systems” is a better starting point than a feature list.
2. Name a business owner who makes decisions and is available every week.
3. Release small and early. A usable first version in weeks, then improvements based on real use.
4. Build on standard, well-supported technology so others can maintain it.
5. Plan for security from the start: access control, logging, data protection.
6. Measure the result against the problem: hours saved, errors reduced, customers served.
7. Budget for maintenance. Software that isn’t maintained becomes a risk.

Web app, mobile app or both?

Most business tools start as web applications because they work on any device and are easier to update. Add a mobile app when people need offline access, device features such as the camera or location, or frequent use on the move. See how to build a web app, the business case for a mobile app and PWA or native app.

Where to start

Before building anything, map the process and check whether connecting what you already own solves it. Our Systems & Integrations work does exactly that, and builds custom tools only where they earn their place. Tell us where you’re stuck.