Build or Buy? Custom Software vs. a Subscription Tool for Your Business

A developer telling you that you need custom software has a conflict of interest, and you should know that before you read anything else we write. We build custom systems. We also tell people, regularly, to buy something off the shelf instead. This article is how we decide which advice to give.

The short version: buy by default, and build only when buying has a specific, nameable cost that building would remove. The rest of this piece is how to tell the difference.

Start by assuming you should buy

For most jobs a business needs done, someone has already built a good tool. Email, accounting, payroll, scheduling, a point-of-sale system, a basic online store: these are solved problems, and the products that solve them have been tested by thousands of businesses and improved for years. Rebuilding any of them is almost always a mistake, however tempting the idea of owning your own version.

A subscription tool also moves a great deal of work off you. Someone else keeps it secure, fixes it when something outside it changes, and adds features without being asked. That is real value, and a custom system has to be worth more than it to justify itself.

Five questions that decide it

If the honest answers to those are 'no, no, no, no and a long time,' a build is worth discussing. If any one of them is a clear 'yes,' stay with the tool.

The arithmetic, with made-up numbers

Here is a comparison using numbers invented for the example, so you can see the shape of it and swap in your own. Suppose a scheduling and job-tracking tool costs $40 per user per month, and you have eight people using it. That is $320 a month, or $3,840 a year. Over five years, if nothing changes, it is $19,200.

Now suppose a custom version costs $6,500 to build and $99 a month to host and look after. Over the same five years that is $6,500 plus $5,940, or $12,440. On those numbers the custom build is cheaper after about two and a half years, and the gap widens every year after that.

Before you get excited, notice what the sums leave out, because they cut both ways.

The point is not that custom wins. The point is that the comparison is a few lines of arithmetic you can do yourself, and any developer who will not help you do it is selling, not advising.

When to buy

When to build

The middle path almost nobody mentions

The choice is not binary. Between a subscription tool and a full custom system sits a third option: a smaller piece, built to do one job well, that connects to what you already use. Online ordering, direct booking and a customer login are all examples, and you can see the options on the add-ons page. Many of the systems we build began life as an add-on that proved its worth, and then grew.

Starting there has a practical advantage beyond cost. You learn how your customers actually behave before you commit to the larger design. We would rather you spend a modest amount to find out than a large amount to guess.

A warning about reinventing commodity software

The most common mistake is not building too little. It is building something that already exists. If a developer proposes a custom accounting system, a custom email platform or a custom payroll tool for a small business, the right response is to ask why. The answer is almost never good enough.

Build the parts that are specific to you: how orders are dispatched, how an auction runs, how your ambassadors are rewarded. Buy everything else, and make sure the two can talk to each other.

What to do next

If you are weighing this up, write down three numbers: what the tool costs now, what it will cost if the business doubles, and how many hours a week your team spends working around it. Those three tell you most of what you need.

Then talk to us, or to anyone you trust, and ask the question that matters: what would you do in my position? Our custom systems each list who they are good for and, just as importantly, who they are probably not for. If a tool you can already buy would serve you better, we will say so.