Product guide
How to scope an MVP you can build in about eight weeks
To scope an MVP, write down the single riskiest assumption your business depends on, map the shortest path a user takes to prove or disprove it, and build only the features on that path. Prioritise everything else with MoSCoW, keep Must Haves to roughly 60 per cent of the effort, and decide your success metrics before writing any code.
- Goal
- Test one risky assumption
- Typical build
- 6 to 10 weeks
- Must Haves
- About 60% of effort
- Decide first
- The metric that means 'it works'
Updated · By the Codebrix team
What an MVP is for
A minimum viable product is not a cheaper version of your full product. It is the smallest thing real users can use that tells you whether your main bet is right. That bet is usually one of three: people have this problem badly enough, they will use your way of solving it, or they will pay for it.
Scoping goes wrong when founders start from a feature list and trim it. It goes right when you start from the question you need answered and add only what is needed to answer it. Everything in this guide follows from that.
Map the core loop
The core loop is the shortest sequence a user goes through to get value from your product. Write it as five to eight steps. For a hypothetical booking app for independent physiotherapists, it might be:
- Physio signs up and sets weekly availability.
- Physio shares a booking link with patients.
- Patient opens the link and picks a slot.
- Patient enters name and phone number and confirms.
- Physio gets notified and sees the booking in a list.
- Patient gets a reminder the day before.
Every feature you are considering either sits on this loop or it does not. Payments, reviews, multi-clinic support, analytics dashboards and a mobile app for physios are all off the loop for version one. They may matter later. They do not help you learn whether physios will share the link and patients will book through it.
Prioritise with MoSCoW
MoSCoW sorts every requirement into Must Have, Should Have, Could Have and Won't Have this time. The Agile Business Consortium, which maintains the method within its DSDM framework, recommends keeping Must Have effort to typically no more than 60 per cent of the total, with a pool of around 20 per cent Could Have effort as contingency. That spare capacity is what lets you hit a date when something takes longer than planned.
| Priority | Meaning | Example items |
|---|---|---|
| Must Have | Without it the loop breaks or the product is unsafe | Sign-up and login, availability set-up, public booking page, booking confirmation, physio's booking list, SMS or WhatsApp reminder |
| Should Have | Painful to leave out but there is a workaround | Reschedule and cancel by the patient, basic email notifications, simple admin view for you |
| Could Have | Nice, and first to drop if time runs short | Custom booking page colours, calendar sync, booking notes |
| Won't Have this time | Agreed out of scope for this release | Online payments, reviews, multi-location clinics, native physio app, reports |
Estimate effort for each item with your developers, then check the split. If you are outsourcing the build, our guide to outsourcing software development to India covers how to set up that estimate and contract. If Must Haves are 85 per cent of the effort, you have no room for surprises. Move items down or simplify them, for example a reminder by email before you build WhatsApp.
What to cut, and what never to cut
Usually safe to cut or fake
- Admin dashboards: use the database console or a spreadsheet export for the first months. A proper internal dashboard can come with your SaaS build once usage justifies it.
- Payments: invoice by hand or use a payment link until you know people will pay.
- Native apps on both stores: a responsive web app often tests the same assumption faster.
- Settings and preferences: pick sensible defaults and let users ask for options.
- Onboarding tours, animations, dark mode and multiple languages, unless your audience requires them.
- Automations: do the task manually behind the scenes and automate it once it repeats.
Never cut
- Secure login and sensible handling of personal data. A leak at MVP stage can end the company.
- Product analytics on every step of the core loop. Without it you cannot answer your question.
- Error tracking and backups, so you find problems before users report them and can recover data.
- A privacy policy and terms, reviewed by your legal adviser for the markets you launch in.
- A way for users to reach you: a support email or chat, monitored daily.
A sample eight-week plan
This is how an eight-week MVP can be laid out with weekly demos. Your timings will depend on scope, design complexity and how fast feedback comes back.
| Week | Focus | What you see at the end of the week |
|---|---|---|
| 1 | Discovery: hypothesis, core loop, MoSCoW list, metrics, technical plan | Signed-off scope document and success metrics |
| 2 | Design: wireframes, then key screens in Figma | Clickable prototype of the core loop |
| 3 | Design finish; project set-up, database, authentication | Approved designs; login working on staging |
| 4 | Core loop part one: availability, booking page | A patient can see open slots on staging |
| 5 | Core loop part two: booking, confirmation, physio's list | A full booking works end to end on staging |
| 6 | Reminders, Should Haves, analytics events | Reminders sending; events visible in analytics |
| 7 | Testing on real devices, fixes, performance, legal pages | Release candidate; your own test with five friendly users |
| 8 | Launch to first real users; monitoring; quick fixes | Live product and a dashboard of your metrics |
If week 1 shows the Must Haves will not fit in eight weeks, change the scope, not the deadline. A date that keeps moving is how MVPs turn into year-long projects.
Metrics to watch after launch
Pick three to five numbers before launch and write down the value that would count as a pass. Deciding afterwards makes every result look like success.
| Metric | What it tells you | Example for the booking MVP |
|---|---|---|
| Activation rate | Share of sign-ups who reach the first moment of value | Physios who set availability and share their link |
| Core action volume | Whether the loop is actually used | Bookings made per active physio per week |
| Retention | Whether value lasts beyond the first try | Physios still getting bookings in week 4 |
| Time to value | How long until a new user benefits | Hours from sign-up to first booking received |
| Willingness to pay | Whether the business works | Physios who agree to a paid plan when asked |
- Talk to at least five users in the first two weeks. Numbers tell you what; conversations tell you why.
- Watch where users drop out of the core loop in your analytics funnel.
- Keep a single list of feature requests and count how many users ask for each one.
- Review the metrics weekly and decide: keep going, change the loop, or change the bet.
Keep scope from creeping during the build
- Put the Won't Have list in the scope document. It is as important as the Must Have list.
- Any new idea goes on a parking list, not into the current sprint, unless something of equal effort comes out.
- Review the MoSCoW list at every weekly demo, with the effort split visible.
- Have one person on your side who can say yes or no to changes the same day.
Codebrix scopes MVPs this way. Product builds are custom-quoted and usually take 6 to 16 weeks; our indicative range for a web application is ₹1,50,000 to ₹6,00,000 ($5,500 to $22,000), and for an iOS and Android app ₹2,50,000 to ₹9,00,000 ($9,000 to $33,000). The fixed quote follows a discovery call. If you are choosing a mobile stack, see React Native vs Flutter and the mobile app cost guide, or read about our MVP development service.
Frequently asked questions
How long should it take to build an MVP?
Many software MVPs can be built in six to ten weeks once the scope is fixed, including design and testing. If your Must Have list needs much longer, that usually means the scope is testing several assumptions at once. Narrow it to the riskiest one and build the rest after you have evidence.
What is the MoSCoW method?
MoSCoW sorts requirements into Must Have, Should Have, Could Have and Won't Have this time. The Agile Business Consortium recommends keeping Must Have effort to typically no more than 60 per cent of the total, with about 20 per cent Could Have effort as contingency, so the release can still land on time if estimates slip.
Should my MVP be a mobile app or a web app?
If your users can do the core loop in a browser, a responsive web app is often faster to build, needs no store review and is easier to change daily. Choose a native or cross-platform app when the loop depends on push notifications, the camera, offline use or other device features, or when your users expect an app.
What features should never be cut from an MVP?
Secure authentication, careful handling of personal data, analytics on every step of the core loop, error tracking, backups, legal pages reviewed by your adviser, and a way for users to contact you. Cutting these saves little time and can either harm users or leave you unable to learn anything from the launch.
How do I know if my MVP worked?
Before launch, write down three to five metrics and the value for each that would count as a pass, such as activation rate, weekly core actions per user and week-four retention. After launch, compare against those numbers and talk to users. If you set targets after seeing results, any outcome will look like success.
How much does an MVP cost to build?
It depends on the number of user types, screens, integrations and platforms. As an indication, Codebrix's range for a web application is ₹1,50,000 to ₹6,00,000 ($5,500 to $22,000) and for an iOS and Android app ₹2,50,000 to ₹9,00,000 ($9,000 to $33,000). These are indicative; the fixed price follows a discovery call.
Sources
Related
MVP development
MVP development: a narrow first product, shipped in 6 to 12 weeks
MVP development company that cuts scope first, then ships a web or mobile MVP in 6–12 weeks with analytics in place. Fixed-price milestones, your code.
Read moreSaaS development
SaaS development with multi-tenancy, billing and admin built in from the start
SaaS development company building multi-tenant products with Stripe or Razorpay billing, onboarding, admin and analytics. Milestone pricing, your code.
Read moreCost guide
What does mobile app development cost in 2026?
Mobile app development cost broken down: build drivers, our indicative ranges, Apple and Google store fees, recurring costs and how to get a fixed quote.
Read moreComparison
React Native vs Flutter: choosing a framework for your app
React Native vs Flutter compared on UI, language, performance, hiring and cost, with a decision list and when to skip both for native apps.
Read moreWeb applications
Custom web application development for portals, dashboards and internal tools
Custom web application development for portals, dashboards and internal tools: roles, Postgres, integrations and tests. Fixed price, code in your name.
Read moreB2B & SaaS
A B2B software development company for portals, dashboards and integrations
B2B software development company for customer portals, dashboards, integrations, SSO and audit logs, built so your SOC 2 preparation has evidence to use.
Read more