Why the Cheapest Quote Is Usually the Most Expensive
A low quote is not a discount. It is usually a different scope, wearing the same name — and the difference arrives later, as a change request.
You have three quotes for the same project. They differ by a factor of three.
The obvious conclusion is that one supplier is expensive and another is competitive. The more likely explanation is that you are not looking at three prices for the same thing. You are looking at three different interpretations of an ambiguous brief, and the cheapest one is usually the narrowest.
The difference does not disappear. It arrives later, as a change request, at a point when you have no leverage left.
This is a companion piece to our guide on what custom software actually costs. Here we are concerned with a narrower question: how to read the quotes in front of you.
A low number is a scope statement
When a supplier quotes low, one of four things is usually true.
They have scoped less than you think. Your brief said "user management." One supplier read that as login and password reset. Another read it as roles, permissions, audit logging, session management, and an admin interface to configure it. Both answered your brief honestly. Only one answered the question you meant.
They have left out the work that is not visible in a demo. Quality assurance is the usual casualty. Software QA analysts earn a mean annual wage of $111,490 in US labour statistics — meaningfully cheaper than developers, but not nothing. (BLS) A quote with no QA line is not a cheaper quote. It is an incomplete one, and the testing still has to happen — it just happens in production, performed by your users.
Security review, accessibility, documentation and deployment tooling disappear the same way.
They have assumed a capacity that does not exist. DORA's research found that even elite-performing engineering teams report spending only about 50% of their time on new work — the rest going to unplanned work, rework, defects and support. Low performers report 30%. (DORA)
A quote built on the assumption that five developers will deliver five developers' worth of features is out by roughly half before work begins. That gap does not show up in the proposal. It shows up in month four.
They are buying the relationship. Quote low, win the work, recover margin through change requests. This is a recognised commercial strategy rather than an accident, and the tell is a contract where the scope is loose but the change process is very well defined.
The comparison problem
Here is the harder issue: quotes are not comparable unless somebody makes them comparable, and that somebody has to be you.
The UK Cabinet Office, which buys more technology than almost any organisation in the world, puts the underlying principle plainly in its own guidance to its own buyers:
"The key component of fixed price has to be 'fixed scope'. Floating or variable scope is not suitable for fixed pricing."
(Cabinet Office risk allocation and pricing guidance)
If your scope is not fixed — and for most genuinely new software it cannot be, because you will learn things — then a fixed price is not a price. It is a bet, and the supplier has priced their side of it. Either they have added a risk premium, in which case you are paying for uncertainty you may not encounter, or they have not, in which case the first real surprise becomes a commercial dispute.
The UK National Audit Office documented where this ends. Across five large government digital programmes, supplier arrangements contributed to delays of at least 29 years and more than £3 billion in cost increases, at least 26% of the original forecast — figures the NAO notes come in varying formats and price bases. Its diagnosis: procurement frameworks "geared to buying inputs rather than outcomes." (NAO, January 2025)
That is what happens at national scale. The same physics operate on a £60,000 project.
Four checks
Normalise the scope before you compare the numbers. Write down every assumption each supplier made, side by side. Where they differ, go back and ask. You will often find the gap between the cheapest and dearest quote is not margin at all — it is six features and a security review.
Ask what is excluded, explicitly. Not "what's included" — suppliers answer that question well. Ask what a reasonable client might expect that this quote does not cover. The quality of the answer tells you a great deal.
Price year three, not year one. Hosting that grows with usage, dependency updates, support, and the change the business will need. Of the more than $100 billion the US federal government spends annually on IT and cyber-related investments, agencies have typically reported spending about 80% on operations and maintenance of existing IT rather than on building anything new. (GAO) That is a portfolio figure for an old estate, not a rule for your project — but the direction is right, and the line belongs in your budget.
Read the change process before the price. If the scope is vague and the change mechanism is precise, you have learned how this contract is intended to make money.
What a fair quote looks like
It shows its assumptions — team composition, seniority mix, duration, rate — so you can argue with them. It includes the unglamorous lines. It states what would have to go wrong for the number to break, and what checkpoint would catch that early. And it prices what happens after launch.
That quote will often not be the lowest. It will usually be the closest to what you actually pay.
If you are comparing proposals and want help normalising them before you decide, we are glad to look — including at proposals that are not ours.
Kaizen Spark Tech designs and delivers software, AI, automation and digital infrastructure for businesses and institutions. Every statistic here is linked to its original published source.