Bespoke Software: When Off-the-Shelf Stops Working

The default advice is to buy rather than build, and it is correct far more often than the software industry likes to admit. An off-the-shelf product has been tested by thousands of companies, someone else pays for its maintenance, and it works on the first day.

But there is a point where continuing to buy costs more than building, and companies usually pass it long before they notice. Here is how to tell.

Signal one: the spreadsheet beside the system

There is a product you pay for, and beside it there is a spreadsheet that somebody maintains by hand, because the product cannot do one specific thing your business needs.

That spreadsheet is the specification for the software you actually need. It is also a liability, because it lives on one laptop, has no audit trail, and breaks silently when the person maintaining it is away.

One spreadsheet is normal. Three is a signal. When people start building tooling around the tool, the tool is no longer fitting.

Signal two: paying for seats that do nothing

Per-seat pricing works well when everyone uses the product the same way. It works badly when you have forty people who need to do one small thing occasionally and five people who use it properly.

At a certain headcount, the annual cost of licensing people who touch two screens exceeds the cost of building those two screens. Do the arithmetic honestly, including maintenance, and the crossover point is often closer than it feels.

Signal three: the process bends around the software

This is the expensive one and the hardest to see, because it happens gradually.

The product has an opinion about how work should flow. Your business had a different one. Over a couple of years, the business quietly reshaped itself to match the software, and now people do things in an order that makes no commercial sense because that is what the system allows.

The cost here is not a licence fee. It is the accumulated inefficiency of every person following a worse process, every day, forever.

Signal four: the thing you cannot buy is the thing you sell

If your competitive advantage lives inside a product every competitor can also purchase, it is not an advantage.

This is the one genuine case for building without much hesitation. The part of your business that is actually distinctive should not be constrained by what a vendor decided to prioritise.

The honest costs of building

Bespoke software is not free after launch. Anyone who quotes it as a one-off number is either inexperienced or hoping you will not ask.

You are taking on:

• Maintenance. Dependencies age, platforms change, browsers update.
• Knowledge. Somebody has to understand it. If that is one person, you have created a new single point of failure.
• Hosting and operations. Real, ongoing, and easy to forget in the business case.

A reasonable rule of thumb is that annual upkeep runs at a meaningful fraction of the original build. If the business case only works when upkeep is zero, it does not work.

The middle path most people skip

You do not have to choose between buying everything and building everything.

The pattern that works most often is buying the commodity parts, accounting, email, storage, and building only the layer that is genuinely specific to you, connected to the rest through integrations. You get the maintenance advantages of bought software where it does not matter, and control where it does.

That is how most of our custom development engagements are actually shaped, and the connective layer between bought and built systems is frequently better served by automation than by writing more application code.