By Quick answer: What is an MVP (minimum viable product)?
- MVP = minimum viable product: the smallest thing that tests whether users want your product.
- It is a learning tool, not a cheap version 1. The goal is evidence, not features.
- It does not have to be an app: Dropbox's MVP was a three-minute video.
- A software MVP in India usually costs ₹3–15 lakh and takes 6–12 weeks; a no-code or manual MVP can cost far less.
Key takeaways
- An MVP tests one risky assumption. If you cannot name the assumption, you are building a small product, not an MVP.
- The cheapest MVP that produces real evidence wins. Code is often not the first step.
- 'Viable' matters: an MVP must actually solve the problem for some users, or the test is meaningless.
- Decide the success metric before launch, so the result cannot be argued away.
- Each MVP should end in a decision: persevere, pivot or stop.
"MVP" may be the most misused term in startups. Founders use it to mean "the first version", agencies use it to mean "the cheap package", and investors use it to mean "something we can try". The original idea is sharper, and far more useful: an MVP is an experiment. Its job is not to impress anyone. Its job is to tell you, as cheaply as possible, whether you are right.
The definition, in one sentence
Eric Ries, who popularised the term, defined the minimum viable product as the version of a new product that lets a team collect the maximum amount of validated learning about customers with the least effort1. Put simply: the smallest thing you can put in front of real users that tells you whether they want what you are building.
Minimum
Only what is needed to test your riskiest assumption. Every extra feature delays the answer.
Viable
It must genuinely work for some users. A broken or useless MVP proves nothing except that broken products fail.
Product
Something people use or buy, not a survey about whether they might. Behaviour beats opinion.
Famous MVPs that were not really apps
The best MVPs are often not software at all.
| Company | The MVP | What it tested |
|---|---|---|
| Dropbox | A three-minute demo video aimed at early adopters | Would people want seamless file sync? The video drove hundreds of thousands of visitors to the waiting list2 |
| Zappos | Photos of shoes from local stores; the founder bought a pair only after an online order | Would people buy shoes online without trying them on?4 |
| Airbnb | Air mattresses in the founders' flat during a sold-out conference | Would strangers pay to stay in someone's home?5 |
Notice the pattern. Each MVP tested one leap-of-faith assumption — the belief that, if wrong, kills the business — with the least possible building.
Types of MVP, from cheapest to most expensive
- Landing page or video MVP — describe the product and measure sign-ups or pre-orders. Costs days. Tests demand.
- Concierge MVP — deliver the service manually to a handful of customers. Costs your time. Tests whether the outcome is valuable.
- Wizard of Oz MVP — the user sees a product; behind it, humans do the work. Tests the experience before automating it.
- No-code MVP — build the core flow with no-code tools. Costs weeks and a small subscription. Tests real usage.
- Single-feature software MVP — a real app doing one job well. Costs ₹3–15 lakh in India. Tests retention and willingness to pay.
Henrik Kniberg's well-known illustration makes the point visually: instead of delivering a car one wheel at a time, deliver a skateboard, then a scooter, then a bicycle — something usable at every step that lets you learn what the customer actually needs3. He even suggests calling the first step the earliest testable product, which is often a clearer name.
MVP vs prototype vs version one
| Prototype | MVP | Version one | |
|---|---|---|---|
| Question it answers | Can it work? How should it work? | Do people want it? | Can we serve the market well? |
| Who uses it | A few testers, often in a session | Real users, in real life | Everyone you can reach |
| Success looks like | Users complete tasks, find it clear | Users adopt, return or pay | Growth, retention, revenue |
| Typical cost | Days to weeks of design time | ₹0 to ₹15 lakh | ₹15 lakh and up |
How to build an MVP in six steps
- Write down your riskiest assumption. "Small clinics will pay ₹999 a month to stop missing appointments" is testable. "People will love it" is not.
- Choose the cheapest test that produces real behaviour — a pre-order, a sign-up, repeat use — not opinions.
- Set the success criterion before you launch: a number and a deadline.
- Build only what the test needs. Cut onboarding polish, settings and edge cases unless the test depends on them.
- Put it in front of the right users — people with the problem, not friends being kind.
- Decide: persevere, pivot or stop. An MVP that ends without a decision was just a small product.
What an MVP costs in India
| MVP type | Typical cost | Time |
|---|---|---|
| Landing page, video or survey-plus-pre-order | ₹0–50,000 | 1–2 weeks |
| Concierge or Wizard of Oz | Mostly your time | 2–6 weeks |
| No-code app | ₹50,000–3 lakh | 2–6 weeks |
| Software MVP, one core flow, cross-platform | ₹3–15 lakh | 6–12 weeks |
The software figures match the "simple MVP" band in our app development cost guide. For a deeper look at what to include and leave out, see MVP development cost for startups.
Why simple first works
Gall's law observes that a complex system that works is invariably found to have evolved from a simple system that worked — and that a complex system designed from scratch rarely works at all. It is the engineering case for the MVP. Read more about Gall's law in our Laws of UX library.
Key terms, defined
Validated learning — evidence from real user behaviour that confirms or rejects an assumption. Leap-of-faith assumption — the belief that, if false, makes the business fail. Pivot — a change of strategy based on what you learned. Concierge MVP — delivering the service manually before building software. Wizard of Oz MVP — a product front end with humans doing the work behind it.
Frequently asked questions
What is an MVP in simple words?
An MVP is the simplest version of your product that real people can use, built to find out whether they actually want it. Instead of building everything, you build just enough to test your biggest assumption — then use what you learn to decide what to build next.
What does MVP stand for in startups?
Minimum viable product. The term was popularised by Eric Ries in the Lean Startup movement, who defined it as the version of a new product that allows a team to collect the maximum amount of validated learning about customers with the least effort.
What is the difference between an MVP and a prototype?
A prototype tests whether something can be built or how it should work, usually with a small group and often without real use. An MVP tests whether people want it, with real users doing real things — signing up, paying, coming back. Prototypes answer design questions; MVPs answer business questions.
What are examples of famous MVPs?
Dropbox validated demand with a three-minute demo video before the product was ready. Zappos tested online shoe sales by photographing shoes in local stores and buying them only after an order. Airbnb began with its founders renting air mattresses in their own flat during a conference.
How much does an MVP cost in India?
A software MVP built by an Indian agency or experienced freelancers usually costs ₹3–15 lakh and takes 6–12 weeks. A no-code MVP can cost a fraction of that, and a landing page, video or manual 'concierge' MVP can cost almost nothing beyond your time.
How do you know if your MVP is successful?
Set a measurable success criterion before launch — for example, 40 of 100 invited users complete the core task twice in two weeks, or 10 paid pre-orders. If you hit it, persevere; if you clearly miss it, change the approach or the idea.
Sources
- Eric Ries, Startup Lessons Learned — Minimum viable product: a guide (2009)
- TechCrunch — How Dropbox started as a minimal viable product (2011)
- Henrik Kniberg, Crisp — Making sense of MVP (2016)
- Wikipedia — Zappos (history)
- Wikipedia — Airbnb (history)
- Checked October 23, 2026. Figures that change — fees, salaries, prices — are reviewed every six months; if something is out of date, tell us.
Where to go from here
- MVP development cost for startups — how lean is too lean
- Gall's law — why complex systems that work evolve from simple ones
- App developmentMVP development cost for startups: how lean is too lean?An MVP costs ₹0–3 lakh as a no-code or manual test and ₹3–15 lakh as a real app in India. What to include, what to cut, and when it's too lean.
- How to make an appWhat is a PWA? Progressive web apps explained simplyA PWA is a website that installs and works like an app, offline and with push. How PWAs work, Indian success stories, iPhone limits and when to choose one.
- App developmentFreelancer vs agency vs in-house: who should build your app?Freelancers are cheapest, agencies fastest for version one, in-house best long term. Costs, risks and a decision guide for hiring app developers in India.
- App developmentHow much does it cost to build an app like Uber? We did the mathAn app like Uber costs ₹30–80 lakh to launch in one Indian city and ₹1–2 crore at full scale. Here's the feature-by-feature math, running costs and shortcuts.
- Templates & toolsApp cost calculator: estimate your app development budget in India in four stepsA free app cost calculator for India: pick your app type, platforms, design and team to estimate a build budget, plus GST and yearly running costs.
- How to make an appHow to build a SaaS product from scratchBuild a SaaS product from idea to paying customers: a painful problem, validation, a narrow MVP, auth, billing, multi-tenancy and your first 10 customers.
Get the ideas, not the noise.
Sign up with Google and we’ll send you short UX psychology tips by email — why people tap, stall and leave, and what to do about it. Unsubscribe anytime. Prefer something to keep? Get the free Laws of UX guide as a PDF.
Spend your budget on what users want.
A behavioral UX audit of your app or SaaS product: why users hesitate, stall and leave, and a prioritized fix list. Fixed price, 2–4 weeks.