Answer six questions and get a planning range for what your MVP will cost and how long it will take, compared across three routes: building it yourself with AI tools, a product-leader-led build, or a development agency. No email required to see the number.
Your build
One is an MVP. Three is a product. Five is a two-year roadmap.
ESTIMATED TIMELINE
8–11 weeks
Build it yourself with AI tools
£300–£1,500 in tooling
≈ $380–$1,900 USD
Tooling and model costs only. Your time is the real cost.
Product-leader-led build (me)
£25,000–£35,000
≈ $31,800–$44,500 USD
One person, product decisions included, four to six weeks.
A planning range for UK, US and EU delivery in 2026, not a quote. USD figures are indicative at £1 ≈ $1.27. Scope discipline is what decides where you land in the range.
THE FRAMEWORK
The Crossley Method: idea to first revenue in seven stages
Founders tend to picture MVP cost as the price of writing code. In practice a six-week build divides roughly into a quarter deciding what to build, half building it, and a quarter on the launch surface — pricing, onboarding, analytics and the payment path. That final quarter is the part agencies quote as optional and the part that determines whether launch teaches you anything.
Why scope moves the number more than anything else
Every additional user journey adds around two weeks, and two weeks compounds: more surface to design, more states to handle, more to test, more to explain during onboarding. Two journeys instead of one rarely costs 100% more — it usually costs 130%, because the interactions between them have to work too.
This is why the single most valuable hour in any MVP project is the one spent cutting scope. It is also why the DIY route is not automatically the cheapest: AI tools remove the cost of typing code, not the cost of building the wrong thing quickly. See product validation for how to avoid that before spending anything at all.
Where AI features change the maths
Simple LLM calls add about a week. Retrieval over your own data adds two to three, because the work is in the data pipeline and the evaluation rather than the prompt. Agentic workflows add four or more, and carry real ongoing cost: every action has a token price, so a heavy user on a £30 subscription can be gross-margin negative before anyone notices. Budget for the evaluation set and the cost model, not just the feature.
Hidden costs the quote usually omits
—Hosting, database and model usage — modest at MVP scale, but not zero.
—Design assets, domain, email infrastructure and legal pages.
—Analytics tooling and the time to define an activation funnel properly.
—Post-launch iteration: assume four to six weeks of changes after real users arrive.
—Your own time — usually the largest real cost on the DIY route, and always excluded from the number.
If you want a real quote rather than a range, the MVP development page sets out exactly what a six-week build includes.
MVP cost questions
How much does it cost to build an MVP in 2026?
A focused MVP with one core user journey costs £12,000 to £25,000 when built by an experienced product leader using AI tooling, or £60,000 to £150,000 through a traditional development agency. Building it yourself with AI tools costs £50 to £300 per month in tooling, with your own time as the real expense.
What drives MVP cost the most?
Scope, by a wide margin. Each additional user journey adds roughly two weeks. After scope, the biggest drivers are AI complexity — retrieval and agentic workflows cost far more than simple LLM calls — followed by third-party integrations and the level of design polish.
How accurate is this MVP cost calculator?
It produces a planning range based on typical UK, US and EU delivery in 2026, not a quote. Real projects vary with data complexity, integration quality and how firmly scope is held. Treat the lower bound as achievable with strict scope discipline and the upper bound as realistic if scope moves.
Is it cheaper to build an MVP with AI tools?
The build is roughly three to five times faster with tools like Lovable, Cursor and Claude Code. What does not get cheaper is deciding what to build, judging whether output is correct, and finding customers — which means product judgement is now a larger share of total cost, not a smaller one.
Should I use an agency or a fractional product leader?
Use an agency when the specification is settled and you need build capacity. Use a product-leader-led build when the scope still needs deciding, which is the usual situation pre-product-market-fit — the cost difference is largely the price of that decision-making being outsourced or included.
EXPLORE NEXT
Continue the journey
Every stage from first idea to first revenue, mapped end to end.