Founders ask me what an MVP costs and I have learned to answer a slightly different question, because the one they asked has a useless answer. An MVP costs between nothing and a million pounds. That range is real and it is not helpful to anybody.
The useful question is: what does a given price tell you about the thing you're going to receive? Because a quote is not really a number. It's a compressed statement about how many people will be involved, how much of their time you're buying, and — the part nobody says out loud — how much scope has to be cut for the maths to work. Most quotes answer that last question without meaning to. Learning to read it is the whole skill.
Start with the day rate, because everything else is derived from it
The floor under any build price in the UK is what a competent engineer costs per day. As I write this, the median daily rate for a software developer contract in the UK is £513, based on job vacancies in the six months to 16 August 2026. Someone senior enough to make architecture decisions on their own — which is what you actually need, and I'll come back to why — is closer to £600.
So: eight weeks of a senior contractor is forty working days, and forty working days at £600 is £24,000. That is your reference price. Not because you should pay it, but because every other number you're quoted is a variation on it, and if you don't know it you can't tell which way the variation runs.
Two things fall out of that immediately.
The first is that anything dramatically below it is buying you fewer days, less seniority, or both. That is not automatically bad — fewer days of the right person is often exactly right — but it is never free. Someone has decided what to leave out. You want to know who, and what.
The second is that anything dramatically above it is buying you people you are not going to meet. An agency quote is a day rate multiplied by a team, plus the overhead that team requires: an account manager, a project manager, a designer with 20% allocated, a QA pass, and the salesperson who wrote the proposal. All of those are real jobs. None of them are the person writing your code. At pre-seed, you are paying for coordination overhead that exists because the agency has more than one client, which is a problem you do not have and should not be funding.
What the number is actually buying at each level
Below about £2,000, you are buying a freelancer's spare evenings or an offshore team's junior bench. There are good outcomes here and I have seen them, but they share a property: the founder was technical enough to specify the work precisely and review it properly. If you could do that, you probably wouldn't be reading this. Without it, you are commissioning something you cannot evaluate, from someone with no incentive to tell you it's wrong.
Between about £4,000 and £10,000, you are buying a small number of days from someone senior, on a scope small enough that they will commit to it. This is the band where fixed price starts to make sense, because the work is bounded enough that a competent person can predict it. It is also the band where the cut becomes the product. You are not buying more; you are buying someone's judgement about less. My own Sprints start at £4k and live in this band, so treat that paragraph with the appropriate suspicion.
Between roughly £15,000 and £40,000, you are buying either a longer engagement with one senior person, or a short one with a small team. Both can be right. The question that separates them is whether your build has genuinely parallel workstreams — a mobile app and a backend and a data pipeline that a single person would serialise. If it doesn't, a team is slower than one person, because coordination is not free and you are paying for it twice: once in money and once in weeks.
Above that, you are no longer buying an MVP. You are buying the first version of a product, which is a legitimate thing to buy and a different purchase. If someone quotes you £80,000 for a pre-seed MVP, they have either misunderstood what you asked for or understood exactly what you asked for and think you'll say yes.
The trap: buying days instead of decisions
Here is the thing that costs pre-seed founders the most money, and it never appears as a line item.
What you are short of at pre-seed is not hands. It is decisions. Which of these eleven features is the one that proves the thesis. Whether the model is good enough. Whether this can be a thin wrapper for now and a real pipeline later. Whether the thing you are about to spend six weeks on is the thing that matters.
If you buy the cheapest available hands, you have bought execution capacity for decisions you cannot make. You will get exactly what you asked for, which will be wrong, and you will find out in week five. The cheap build was not cheap; it cost you the money and the six weeks and the answer you were trying to buy.
This is why I keep saying "senior enough to make architecture decisions alone". It is not a status claim. It is the specific thing you are paying the premium for, and if you're not getting it, the premium is wasted.
How to read a quote in about four questions
Ask what gets cut. Not "is this everything I asked for" — ask them to name the three things they would remove first and why. Someone who has done this before will answer immediately and with opinions. Someone who hasn't will say it depends on your priorities, which means they intend to build all of it and find out in week three.
Ask who writes the code and whether you will meet them. If the answer involves the word "team" and no names, the price includes people whose job is to manage the people whose job is to build.
Ask what happens on the day it ships. Where does it deploy, who owns the repository, what documentation exists, what happens if you never speak to them again. An MVP you cannot maintain or hand to someone else is not an asset, it is a demo with a hosting bill.
Ask what the price is fixed against. "Fixed price" and "fixed scope" are the same sentence or neither is true. If the scope is a conversation and the price is a number, the scope will move and the number will follow it.
The honest summary
A pre-seed MVP that answers one specific question, built by one senior person, on a scope small enough to commit to, costs somewhere in the low thousands to low tens of thousands. Where it lands inside that has less to do with the ambition of your product than with how much you are willing to cut — which is a decision you make, not one you buy.
And if the number you're quoted is far outside that range in either direction, the number isn't the problem. It's telling you something about the shape of the engagement, and the shape is what you should be arguing about.
If you want to test that against a specific build, the scoping call is thirty minutes and there's no pitch in it. Worst case you leave with a clearer idea of what to cut, and go and get it built by someone else.
Sources: ITJobsWatch — UK Software Developer contract rates.