---
title: "MVP development: from prototype to product | QMates"
source: https://qmates.tech/en/services/software-product-development/mvp-development
description: "Your MVP wins the demo and cracks with the first customers? MVP development stalls on an undefined question, deferred tests and oversized architecture."
updated: 2026-11-21
---

MVP development

# The prototype works, the product doesn’t yet

The limits of an MVP show up with the first customers, but they start earlier, at the demo: the question the prototype was meant to settle was left implicit, tests and pipeline were put off until later, and the people who built it will hand the product to others. The shortcuts of the demo raise the price of tests, and without tests the architecture stays the one from the demo.

## Does this sound familiar?

-   The demo convinces, then every time a real customer uses the product something nobody expected turns up.
-   The prototype was built fast for a presentation, or on a no-code tool, and now it has to become the product.
-   Every change breaks something somewhere else, and the team is no longer sure it’s worth carrying on from this base.
-   The work has been going on for some time, and it isn’t clear which question the prototype was meant to settle, or whether it did.
-   The code was written by a vendor or by someone who is no longer here, and the team doesn’t know how to carry it forward.

## Why doesn’t the prototype become a product?

It all starts with a question that was never defined. With no criterion for saying when the prototype has answered, it keeps growing and pushes tests and pipeline to an ‘afterwards’ that never comes. An oversized architecture is where all this shows up in the code.

1.  ### The question the prototype was meant to settle stayed implicit

    Without a precise question and a criterion for saying whether the answer is yes, the prototype keeps growing and decides nothing. Scope widens with every meeting, and what should have been a test becomes a half-built product, made in the rush of a demo.

    Feeds into

    -   
    -   

    [

    Handled by

    Product Development & Open Innovation](https://qmates.tech/en/services/software-product-development)

2.  ### Tests and pipeline put off until later

    In the prototype, tests and a release pipeline were a luxury, and afterwards there was never a moment to add them while customers waited for features. Without them every release can break something that worked, and the team puts releases off for as long as it can.

    Feeds into

    -   

    [

    Handled by

    Technical Coaching in XP and TDD](https://qmates.tech/en/services/engineering-transformation/technical-coaching-xp-tdd)

3.  ### An architecture oversized for the stage

    The shortcuts taken for the demo stay in the code, and changes cost more with every feature. A distributed system or a fashionable stack chosen before having any customers can do the same, slowing down a product still looking for its shape.

    [

    Handled by

    Product Development & Open Innovation](https://qmates.tech/en/services/software-product-development)

4.  ### The reasons behind the decisions stayed with whoever wrote it

    If the prototype was written by an external vendor or a single person, the reasons behind the decisions stay with them. The people who carry the product forward do so without knowing them, and every intervention costs more than expected.

    Feeds into

    -   

    [

    Handled by

    Product Development & Open Innovation](https://qmates.tech/en/services/software-product-development)

- [The service that tackles it](https://qmates.tech/en/services/software-product-development)

## What changes when the prototype becomes a product

Every sprint ends with a release to production, and changes go out small, covered by tests and separated into modules. When users ask to change course, the product changes course without the team having to stop and put the rest back together.

In the pipeline every release passes through the tests, and new customers come on board once production is stable. The cost of a change stays predictable while customers grow.
