When you should not build custom software

This is an odd thing for a company that builds software to write, but here it is: sometimes you should not build anything at all. Custom software is the right answer more rarely than the people selling it would like you to believe, and knowing when to walk away will save you more than any discount.

So before you commission anything, run your idea past these.

Does something off the shelf already do 80 percent of this?

There is a vast world of ready made tools for the common jobs: accounting, scheduling, email, invoicing, simple shops, project tracking. If a product already does most of what you need for a monthly fee, building your own version is usually a way of spending fifty thousand pounds to slightly prefer the buttons.

The honest test is whether the missing 20 percent genuinely matters to your business, or whether you just like the idea of something that is yours. Liking the idea is not a business case.

Is the thing you want actually a process problem?

A lot of software requests are really frustrations in disguise. The work is slow, or messy, or error prone, and software feels like the fix. Sometimes it is. Often the same frustration can be solved by changing how the work flows, with tools you already pay for. Software cannot rescue a broken process, it just makes the broken process faster.

Will it still matter in a year?

Custom software is a commitment. It has to be run, maintained and occasionally fixed. If the need is a one off, or a passing experiment, a lightweight tool or even a good spreadsheet may be the grown up choice. Not everything deserves to become a system.

When it genuinely is worth building

To be fair to our own trade, there is a clear case for custom software, and it is a strong one. Build when the thing is core to how your business runs or makes money, when no off the shelf tool fits without bending your business out of shape, and when getting it right is a real advantage rather than a nice to have. A portal your customers live in. A platform your operations depend on. A capability your competitors do not have.

In those cases custom is not indulgence, it is exactly right, and doing it well pays for itself.

The point

The reason to be honest about this is simple. A partner who will only ever tell you to build is not advising you, they are billing you. The ones worth trusting are the ones who will occasionally talk you out of the very thing they sell, because they would rather keep the relationship than win the invoice.

So ask the awkward question out loud: should we even build this? If the answer is no, you have just saved a small fortune. If the answer is yes, you now know why, which is the best possible way to start.