Every business under pressure faces the same tempting maths. There is a thing that needs building: a system, an automation, a website. There are two ways to get it. The fast, cheap way ships in weeks and costs very little. The proper way takes longer and costs more. When money is tight and the pressure is on, the cheap way wins almost every time. Often that is the right call. Sometimes it is a loan against next year, taken out at a rate nobody quoted.
Cheap and fast are real, until they are not
The appeal is not irrational. A quick build genuinely does cost less to produce and genuinely does ship sooner. For plenty of things, that is the correct trade. Not everything needs to be engineered to last a decade. A throwaway landing page for a two-week campaign should be thrown away, not gold-plated.
The trap is applying throwaway economics to something the business will actually depend on. When a quick fix becomes load-bearing, the corners that were cut to make it cheap turn into the reasons it fails later. The saving was real on the day. The bill just had not arrived yet.
Where the deferred cost hides
A build shipped without rigour accumulates cost in places that do not show up until later:
- Tech debt. Shortcuts pile up until every new change takes longer and risks breaking something else. The system becomes slow and frightening to touch.
- Security exposure. Cheap builds skip the unglamorous work of doing authentication, data handling and permissions properly. The gap is invisible until it is not.
- Compliance and privacy risk. Personal data handled without care is a legal and reputational problem waiting for a trigger.
- No documentation. When the one person who understood it leaves, you own a system nobody can safely change.
- Rework. The most expensive outcome of all, where the quick version has to be ripped out and rebuilt properly, so you pay twice and lose the time in between.
None of these are visible in the quote. All of them are real and they compound.
The clue is what happens after launch
The honest test of a build is not how it looks on the day it ships. It is what the next year feels like. A proper build gets quietly cheaper to live with: changes are safe, the thing is understood, new features slot in. A cheap build gets quietly more expensive: every change is a gamble, small requests take weeks and a low hum of anxiety surrounds touching it at all.
Picture a business that commissioned a low-cost system to save money in a tight quarter. Eighteen months on it is paying a premium for every small change, cannot recruit anyone willing to maintain it and is nervously planning the rebuild it thought it had avoided. The cheap option turned out to be the expensive one. It just spread the cost out where nobody was watching.
Deferred build cost is one form of value leaking out of a growing business. Where scaling businesses leak value covers the others, most of which sit nowhere near technology.
How to tell a saving from a deferred bill
The question is not "cheap or expensive". It is "will the business depend on this". If the answer is yes, the corners that make it cheap are the corners that will cost you. A few things separate a genuine saving from a loan:
- Does the build follow basic engineering standards, with security and data handling done properly rather than bolted on later?
- Is it documented well enough that someone other than the original builder can maintain it?
- Is it built to be changed? Or will the first significant change mean starting again?
- Can whoever is building it say what they are trading away to hit the price? Or are the trade-offs invisible?
If nobody can answer those, the low price is not a saving. It is a bill deferred to a year when it will be larger and land at a worse time.
How to price the bill you have already deferred
Deferred cost is measurable, even though it never appears as a line. Four readings.
| Where to look | What to pull | The calculation | What it tells you |
|---|---|---|---|
| Job or ticket system | Work reopened or reworked, against work closed | Reopened ÷ closed = rework rate | The clearest signal that something shipped before it was finished |
| Change records | Median days from a small change being requested to it going live | Median days to change = drag | Rising drag is tech debt arriving, long before anybody calls it that |
| Your systems inventory | Systems with no current documentation, against all systems relied on | Undocumented ÷ total = exposure | How much of the operation depends on one person still being employed |
| Accounting ledger | Spend on rebuilding or replacing something delivered in the last three years | Rebuild spend ÷ original spend = the actual multiple | What the cheap version really cost, once |
The one number to hold
- Rework rate.
Reopened ÷ closed, by system rather than in aggregate - By system, because one badly built thing will otherwise hide inside a healthy average
- Anything trending upward on a system nobody has touched is deferred cost surfacing
The control worth introducing
- Before signing off a cheap option, write down which category it falls into. Temporary and disposable, against something the business will rely on
- Anything relied upon gets documentation as a delivery condition, not as a later phase
Where your number will mislead you
- Rework rate flatters any team that raises a new ticket rather than reopening the old one
- Drag rises for good reasons too, such as more careful release process. Read it alongside incidents
- Rebuild spend is the honest number but the slowest to collect, since it needs three years of history
Build it once, properly, for the things that matter
The pragmatic position is not "always build the expensive way". It is to be honest about which things the business will rely on and build those properly the first time, while being genuinely ruthless about throwing away the things that are meant to be temporary. The waste is not spending money on quality. The waste is spending it twice, because the cheap version of something important always comes back and it never comes back at a convenient moment.
Working out which things your business genuinely relies on is the first job, not the second. The Free Health Check scores six operating areas in about three minutes. How we work sets out how the evidence gets gathered.