Should you buy software or get it built? A plain guide for business owners

Almost every growing business hits this moment. The software you started with worked fine when you were twenty people. Now you are eighty, and it is creaking. Half the real work happens in Excel sheets and WhatsApp groups. Someone says, "Why don't we just get our own software made?" Someone else says, "There are hundreds of products out there. Just buy one."

Both of them have a point. So how do you decide?

Ask a better question

"Buy or build?" treats your business as one big thing. It isn't. It is lots of separate jobs: paying salaries, sending invoices, taking orders, planning production, chasing payments. Some of those jobs are the same in every company. Others are done your own way, and that way may be the reason customers choose you.

So ask this instead, one job at a time: How important is this job to winning and keeping customers? And how different is the way we do it from everyone else?

Payroll is the same almost everywhere. So are accounts, GST filing and email. Plenty of good products already handle these, and there is no point reinventing them. But the way you price a complicated order, plan work across your machines, or get a product approved for a demanding customer might be yours alone.

You have more than two choices

It is not just "buy" or "build". There are really four options, from least effort to most:

1. 01Use it as it comes. Buy a product and change your way of working to fit it. Quickest and cheapest. Ideal for routine jobs.
2. 02Adjust the settings. Most products let you add your own fields, approval steps and reports. Good when your needs are only a little different.
3. 03Add to it. Keep the product you have and build something small around it. For example, a screen for the one step it can't handle, or a link so data flows into it automatically instead of being typed in twice.
4. 04Build it. Have software made around the way you actually work. Worth it when the job is important, done your own way, and nothing on the market fits.

Option 3 is the one people forget. Many businesses don't need to throw out their accounting software or ERP (the big system that runs stock, purchases and sales). They just need the work that happens before and after it organized, and the data to flow in without someone retyping it.

Signs your current software is holding you back

• The real work happens in spreadsheets and messages. The official system only gets the final numbers.
• Only one or two people know the tricks to make the system do what you need. When they're on leave, things stall.
• You pay for extra modules or user licences just to get one process half-working.
• Managers phone around to find out where an order is, because the system can't tell them.
• Every software update breaks something, so you've stopped updating.

One of these on its own doesn't mean you need new software. But if several are true for a job that really matters, you are already paying for the problem every day. It just doesn't show up on an invoice. It shows up as wasted hours, mistakes and delays.

When you should not build

Having software built is a real commitment. Someone on your side has to own it. The people who know the work have to give their time to explain it and test early versions. And like any machine, it needs looking after. So don't build when:

• The job is routine and a well-known product already does it well enough.
• Your process keeps changing. Software tends to lock in whatever you build it around, confusion included.
• Nobody in the business has time to explain the work and try out early versions.
• The real problem is training, discipline or messy data. New software won't fix those.

Compare the real costs

Licence fees are easy to compare, but they are rarely the biggest cost. Look at three to five years and jot down a rough number for each of these:

• Licence or subscription fees as your team grows, including add-ons you need for just one process.
• Setting it up, whether your own staff do it or you pay someone.
• Hours your people spend today on workarounds, double entry and checking one sheet against another.
• Linking products together when they don't talk to each other.
• If you build: the first version, hosting, upkeep and improvements over time.
• Mistakes and delays the current way causes, even if you can only guess at them.

Rough numbers for all of these are more useful than an exact number for one. Writing them down starts the right conversation about where your time and money really go.

A simple way to decide

Take each job that matters and ask two things: is it central to your business, and do you do it differently from most companies? Then:

• Routine and not central (payroll, for example): buy a product and use it as it is.
• Routine but central (your accounts, say): choose a strong product carefully and set it up well. At most, build the links around it.
• Your own way, but not central: add to what you have. A small app or a link between systems often does the job.
• Your own way, and central: this is where software built for you usually pays for itself, and can even give you an edge.

If you do build, do it properly

New software should make life easier than the workaround it replaces. In practice, that means:

• Start from how the work really happens today.
• Get a useful first version into people's hands quickly, instead of waiting a year for everything.
• Make rules like prices and approval limits easy to change without a developer.
• Make sure you own the code, the data and the documents.

Our page on custom business software explains how we work.

Whichever way you go, decide with the process written down and the costs in front of you. If you'd like a second opinion on a particular part of your business, tell us what you're trying to do. We'll tell you honestly if buying is the better answer.