Every founder asks this question and almost every answer is useless. "It depends" is technically true and practically worthless. So here is the version with numbers in it.
We build MVPs for founders across the USA, the Gulf, and Central Asia. The figures below come from projects we have actually shipped, not from a pricing calculator.
The short answer
| What you are buying | Cost | Timeline |
|---|---|---|
| Clickable prototype | $2,000 - $5,000 | 2 - 4 weeks |
| Focused MVP, one platform | $12,000 - $25,000 | 8 - 12 weeks |
| MVP with payments and integrations | $25,000 - $40,000 | 10 - 14 weeks |
| Multi-module platform or v1 | $40,000 - $80,000 | 16 - 24 weeks |
If someone quotes you $4,000 for a marketplace with payments and an admin panel, they are either misunderstanding the scope or planning to renegotiate later. Both end the same way.
Five things that actually move the number
Scope is the obvious one. These are the five that surprise founders.
Number of user types. A product with buyers and sellers costs roughly 60% more than one with a single user type. You are building two products that have to agree with each other, plus the admin view that sits above both.
Payments. Adding a payment flow is rarely just an integration. Refunds, failed transactions, partial captures, held funds, receipts, and reconciliation all become your problem. Budget 50 to 80 hours, not 10.
Real-time features. Chat, live tracking, and video are a different class of engineering. Live video in particular can cost more than the rest of the product combined.
Number of platforms. Web and native mobile at the same time roughly doubles the build. Most MVPs do not need both. Responsive web covers the workflow for the majority of B2B products.
Languages. Adding a second language is not a translation task. Layout, date and number formats, and right-to-left support for Arabic all touch the codebase. Budget two to three weeks.
A real breakdown
Numbers in a range are easy to ignore. Here is a complete estimate from a project we scoped: a B2B marketplace connecting fleet operators with maintenance vendors in Qatar.
| Module | Hours | Cost |
|---|---|---|
| Foundations, environments, CI/CD | 40 | $1,600 |
| Design system and screen build | 90 | $3,600 |
| Accounts, roles and permissions | 50 | $2,000 |
| Vendor directory and matching logic | 70 | $2,800 |
| Job posting and quote flow | 90 | $3,600 |
| Payments with escrow-style hold | 60 | $2,400 |
| Ratings and reviews | 30 | $1,200 |
| Notifications: WhatsApp, SMS, email | 35 | $1,400 |
| Admin panel | 55 | $2,200 |
| QA, UAT and bug fixing | 80 | $3,200 |
| Deployment, documentation, handover | 30 | $1,200 |
| Total | 630 | $25,200 |
Two things are worth noticing.
QA is 13% of the build. Estimates that leave testing out are not cheaper, they just move the cost to the weeks after launch when you are fixing things in production.
The admin panel is 55 hours. Every founder forgets the admin panel. You will need it on day one to approve users, resolve disputes, and see what is happening. It is not optional and it is not free.
The six line items most estimates forget
These appear in month one whether or not anyone priced them.
Payment gateway onboarding. Not the code, the commercial process. In the Gulf this takes three to six weeks and it is on the critical path. Start it before development, not after.
Transactional messaging. WhatsApp Business API approval takes two to four weeks. SMS providers need business verification. If your users are not desk-based, email alone will not reach them.
App store release. One to two weeks of review, rejections, and resubmissions. Apple in particular will find something.
Infrastructure running costs. Hosting, storage, message credits, payment fees. Expect $250 to $450 per month at launch volumes for a product like the one above. This is not a build cost but it starts the month you go live.
Legal content. Terms of service and a privacy policy. Required by both app stores and by payment providers.
Post-launch support. Thirty days is a reasonable inclusion. Anything beyond that is a retainer, and it should be priced before launch rather than negotiated in a crisis.
An estimate that omits these is not a cheaper estimate. It is an incomplete one.
What the region changes
Rates vary by cost of living far more than by capability.
| Region | Senior engineer, blended hourly | Same 630-hour build |
|---|---|---|
| United States | $120 - $200 | $75,000 - $126,000 |
| Western Europe | $90 - $150 | $57,000 - $95,000 |
| Eastern Europe | $50 - $80 | $31,000 - $50,000 |
| Central Asia | $30 - $50 | $19,000 - $31,000 |
The number that matters is not the rate. It is whether the team has shipped something in your category before. A cheap team learning your domain on your budget is the most expensive option on this list, regardless of where they sit.
Fixed price, hourly, or equity
Three models, and the right one depends on how well defined your scope is.
Fixed price works when the scope is genuinely known, which means someone did discovery first. A fixed price quoted from a one-hour call is not a fixed price. It is an opening position.
Hourly is honest for open-ended work but transfers all risk to you. It also creates a quiet incentive problem: the team is paid more when the work takes longer.
Cash plus equity lowers the cash you pay and ties the team to the outcome. It only makes sense if you want a long-term product partner rather than a delivery vendor, and if you have a legal structure that can issue shares. We work this way for a reason: if the product does not work, we lose too.
None of these is universally better. What matters is that the model matches the situation, and that whoever is quoting explains why they picked it.
How to get an estimate you can trust
Four questions. The answers tell you more than the number does.
"Can you break this down by module?" A single figure hides the assumptions. A breakdown shows you what can be cut if the budget is tight.
"What is explicitly not included?" The best answer here is a long one. A partner who names the exclusions has thought about the project. One who says "everything is included" has not.
"What is on the critical path that depends on me?" Payment onboarding, API approvals, content, decisions. If they cannot name these, they have not built this kind of product before.
"What would you cut if I had 30% less budget?" A good partner answers immediately, because they already know which features are load-bearing and which are not.
The uncomfortable part
The most expensive line in most MVP budgets is not a module. It is the three months spent building the wrong thing.
We spend ten days before any build producing a blueprint: scope, architecture, screen designs, and a module-by-module estimate. It costs a fraction of the build and it is the only reliable way to make the number on a quote mean something. Every artefact belongs to the client, including if they take it to another team.
That is not generosity. An estimate you cannot verify is not an estimate, and a blueprint that only works if you hire us would be a sales document rather than a plan.
Working on something and want a real number rather than a range? Tell us about it and we will reply within 48 hours. Our full pricing is published openly.
Frequently asked questions
Written by
Shakhbozbek Usmonov
Founder & CEO, Steppe Venture Builders
