MVP development for startups

If you have a few conversations with potential customers behind you and feel the pressure to finally show something to an investor, a co-founder or yourself, the easiest mistake is to order the full app straight away. Most startups fail because they built something nobody wanted to buy, not because of weak technology. An MVP is meant to protect you from that: the smallest version of the product that lets you check whether the problem exists and whether people will pay to solve it.

Every złoty spent before validation is a bet. A few weeks of building features nobody will use costs more than the code itself, because you also lose time you could have spent talking to customers. We build MVPs as part of our web app offer, but in a different order: first we test the idea on the market, and only then do we design and code.

An MVP answers one question: is this problem painful enough for someone that they will pay to solve it. Anything that does not help answer it waits for a later version. That keeps the first version smaller and cheaper, and you get answers from real users sooner, instead of from friends who will always say the idea is great.

Validation starts with a landing page with a value proposition, a waitlist and a single call to action. We drive traffic to it and count the market signal: how many people leave an email, how many click through to pricing, how many reach the end of a payment test. After a week you have your first numbers.

We run this stage through our strategy and direction service: together we set the hypothesis, the target group and the threshold at which the signal is strong enough to start building. If the idea fails validation, that is a result too. It is cheaper to learn that from a landing page and a test for about PLN 3,000 net than from a finished product for PLN 40,000 net that nobody wants. What happens next, a pivot or shelving the idea, is your call, but you make it on data.

The landing page is half of validation. At the same time you talk to people from the waitlist, and we help you write questions that do not lead the answer. A handful of those conversations can shift discovery priorities more than weeks of debate between founders, so we schedule them right alongside the landing page.

The schedule has a steady rhythm, so from day one you know what is happening in any given week and when you will see results. We work in weekly sprints, and every Friday you get a demo of the working product that you can click through yourself on a preview before anything reaches users.

The pace also depends on you. When approvals and answers come within a day or two, the sprints stay on plan. So at the start we agree on one person on your side who can decide on scope and priorities without calling a meeting.

• Validation (week 1)

  a landing page with a value proposition, a waitlist and a single CTA. We drive traffic and measure the signal

• Discovery (week 2)

  user stories, user flows and feature priorities. What must be in v1 and what can wait

• Design (week 3)

  wireframes, a clickable prototype, user testing and a UI kit

• Build (weeks 4–8)

  Next.js and React, weekly sprints, a demo every Friday, a deploy after every approved change

• Launch (weeks 8–10)

  Product Hunt, beta users, a feedback loop and the first iteration

The hardest decision in an MVP is what not to build. During discovery we write user stories and split features into three groups: essential to test the hypothesis, useful after the first customers, and ones that only look good on a list. Only the first group goes into version one, however tempting the rest may be.

Before we start coding, a clickable prototype comes out of our design that sells service and goes through user testing. Fixing a screen in a prototype takes minutes; in a finished app it takes days, so we catch the places where people get lost before you pay for code. A typical full MVP includes:

• Login and accounts

  email, Google or magic link, with roles only where they are really needed

• Payments

  Stripe with a subscription or a one-off purchase, to test whether people pay

• Dashboard

  the main view where users get what you promised them

• Email and monitoring

  notifications through Resend and errors in Sentry from day one

• Analytics

  tracking of key behaviour, so decisions after launch are based on data

An MVP costs from PLN 12,000 net for validation plus a simple product, up to PLN 40,000 net for a full MVP with login, payments and a dashboard. Validation alone, a landing page with a test, is about PLN 3,000 net. We give the final figure after discovery, because the word “app” is not enough to price a project honestly. Packages and extras, such as push notifications or a CRM integration, are in our pricing.

Three things move the price most: the number of user roles, payments, and connections to outside systems. Each can be added later, so we often recommend starting with a narrower scope. Payments and email are wired up through our integrations and automation service. You get a free initial quote with a written scope and schedule within 48 hours of a call.

We bill by stage and issue a VAT invoice for each one, so you do not pay for everything up front. If you are not sure whether to start with validation alone or go straight to a full MVP, describe the idea in two sentences in the contact form. We reply personally by the end of the next working day, and Kacper Baranowski runs the project from the first call to launch.

After a successful validation, the last thing you want to hear is that version one has to be written again. We build the MVP on a production stack from the start: Next.js and React, a Neon Postgres database with Drizzle ORM, login through NextAuth, and hosting on Vercel with a preview of every change and automated tests. Whatever works after validation, you keep building on.

The app is built through our apps and PWA service, so the MVP works well on a phone from the start. If your users mostly work in the field, we can turn it into a PWA that installs from the browser and works offline. When the product needs a presence in the app stores, the next step is a mobile app.

For you as a founder, the stack has one more advantage. Developers who know it are easy to find, and an investor or a future CTO who opens the repository will see tools they already know. That makes hiring your first developers and talking about the next funding round easier.

Launch is where the real work begins: you watch how people actually use the product and decide what to develop next on that basis. We help with your first launch on Product Hunt, in SaaS founder communities, on Reddit and IndieHackers, then collect feedback from your first users and plan the iteration.

From there you have two paths. We can stay with the product through support and growth: weekly sprints, new features, fixes and monitoring. Or we hand the project over to your team with full documentation; the stack is standard, so a new developer is up to speed in a few hours. Either way, we fix bugs caused by our code free of charge for 30 days from handover, and the code and accounts belong to you from day one.

If your product is mainly a portal for your customers, with subscriptions, roles and reports, see how we build a SaaS client portal. It is often the natural next step after an MVP that has passed validation and started to earn. If all you have so far is the idea, start with validation: describe it in the contact form and we will come back with a suggestion of where to begin.