Here is a number that should make any executive pause before signing off on a custom build: large IT projects run, on average, 45% over budget, 7% over schedule, and deliver 56% less value than predicted. That figure comes from McKinsey and the University of Oxford, who studied more than 5,400 projects. The combined cost overrun was 66 billion dollars — more than the GDP of a small country.
We build custom software for a living. And yet the most valuable advice we give clients is often: don't. The build-vs-buy-vs-customise decision is where most software money is quietly wasted, and the reason is almost always the same — companies default to one answer out of habit instead of deciding on purpose.
Most companies default wrong — in both directions
There are two failure modes, and they are mirror images of each other.
The first is build-by-default. An ambitious team decides that because their business is unique, their software must be too. So they build a CRM, a billing engine, an internal tool — things that thousands of vendors already do well. The Standish Group's CHAOS research found that large software projects succeed less than 10% of the time, while small, focused ones succeed roughly 90% of the time. Building a generic system from scratch is the single most expensive way to acquire ordinary functionality.
The second is buy-by-default, and it is just as costly — it just hides better. The average company now runs around 130 SaaS applications, according to SaaS-management research. Each one solves a real problem on day one. Together they create a sprawl of subscriptions, overlapping features, and integration glue that nobody owns. Buying feels safe because the cost is small and monthly. The total, three years later, is neither.
The hidden cost is never the price on the invoice
Whichever way you lean, the sticker price is the least interesting number. Roughly 65% of a software system's total cost arrives after the initial deployment — maintenance, integration, training, upgrades, and the slow tax of working around things that almost fit.
That single fact reframes the whole decision. A custom build is not a one-time invoice; it is a long-term commitment to maintain something. A SaaS subscription is not a fixed monthly line; it is a price that tends to climb, a roadmap you don't control, and data you may struggle to extract. The honest comparison is total cost of ownership over three to five years — not the quote in front of you today.
A framework that actually decides
Strip away the noise and three questions settle most cases.
1. Is this a core differentiator, or plumbing? If the software is your competitive edge — the thing customers pay you for — build it. If it is plumbing that every business needs (email, accounting, payroll, generic CRM), buy it. Never build what you could buy to run the business; build what makes the business worth running.
2. Does an off-the-shelf tool fit 80% or more, today? If yes, buy it and adapt your process slightly. If it fits 50–80%, customise — configure and extend a proven platform rather than starting from zero. Below 50%, and only if it is core, build.
3. Will the requirement still be true in three years? Stable, well-understood needs are perfect for buying. Fast-moving, genuinely distinctive needs justify building. If you can't answer this one, you are not ready to decide — and that is useful to know before you spend.
Why we sometimes tell clients not to hire us
True Dev has built software since 1997. Long enough to have watched plenty of custom projects that should have been a 40-dollar-a-month subscription, and plenty of subscriptions duct-taped together where one well-built system would have paid for itself in a year.
So when a client asks us to build, our first job is to challenge the premise. Sometimes the answer is a clean custom system because the need is genuinely theirs. Sometimes it is "buy this tool, we'll integrate it, and keep your budget for the part that's actually unique." Honest advice occasionally costs us a project. It has never cost us a client.
If you are weighing a build, buy, or customise decision, start a project with us and we'll help you decide before anyone writes a line of code — or see how we approach software across the whole lifecycle.
