The two at a glance
| Prototype | Production-ready MVP | |
|---|---|---|
| Question it answers | Do people understand this and want it? | Will people use it, return and pay? |
| Who uses it | A handful of test users, with you present | Real customers, without you |
| Data | Sample or fake data | Real data that must be kept safe |
| Built with | Design tools, no-code tools or quick code | Production code, tested and deployed properly |
| Lifespan | Days or weeks, then discarded | Years, extended release by release |
| Success means | You learned something that changes the plan | Customers complete the core job and come back |
| Failure costs | Little | More, which is why the prototype comes first |
Prototype can mean four different things
Part of the confusion is that one word covers several tools. Each tests something different.
| Kind | What it is | What it tests |
|---|---|---|
| Clickable design | Screens linked together in a design tool. Nothing works behind them | Whether people understand the flow and the value |
| No-code build | A working version assembled in a no-code tool | Whether people will use it for a real task, at small scale |
| Technical proof | Rough code for the one hard part | Whether the difficult thing can be done at all |
| Concierge test | The service delivered by hand, with a simple front | Whether people want the result enough to pay |
None of these needs to be good software. Each needs to answer its question as cheaply as possible, which is the whole point.
When each one fits
Build a prototype when
- You have not yet shown the idea to the people it is for
- You are choosing between several directions
- One technical risk could sink the product
- You need something to show investors or partners
- Being wrong now would be cheap, and later expensive
Build a production-ready MVP when
- Prototype tests gave a clear yes
- Customers are waiting to use it for real work
- The product will hold personal or financial data
- People will pay through it
- You intend to keep and extend what you build
What production ready means
Production ready does not mean large or polished. It means the product can be trusted with real use. An MVP can have one feature and still meet every line below.
Production-ready checklist
0 of 8 checkedThe trap, launching the prototype
The costly mistake is in the middle. A prototype works well in a demonstration, someone says it is nearly done, and it is opened to customers. It then holds real data with none of the protections above, and every new feature is built on code that was never meant to last.
Signs it is happening
- The demo is described as ninety percent finished
- Real customer data is being entered into a test tool
- Nobody can say how a release or a backup works
- One person built it and only they can change it
- Fixing one thing regularly breaks another
The remedy is to decide in advance which one you are building, and to say so. A prototype that taught you what to build has done its job. Rebuilding properly afterwards is not waste. It is the plan.
A worked example
From idea to product in three steps
A founder, invented for this example, wants to build a tool that helps small landlords track repairs.
- Step one, a clickable designShown to eight landlords. Six misunderstand the main screen. Two ask for a feature nobody had planned, sending jobs straight to tradespeople.
- Step two, a no-code versionRebuilt around that request and used by five landlords for a month. Four keep using it. Two ask what it costs.
- Step three, a production-ready MVPBuilt in code, with accounts, data separation, payment and the single journey that the tests proved. The no-code version is retired.
Two cheap prototypes changed the product before any production code was written. The MVP that followed was smaller and more certain than the first idea.
What to do next
Write down the one question you most need answered. If it is about whether people want the product, prototype. If it is about whether they will use and pay for it over time, build the MVP. Our comparison of an MVP and a full product build covers the step after that.
A product discovery workshop ends with a tested prototype and a plan, and startup MVP development takes it to production. Or tell us where you are, and we will say which step we think comes next.
Common questions
Can a prototype be turned into the product?
Rarely well. Prototypes skip security, testing and structure in order to be fast, and adding those afterwards is often more work than starting cleanly. What does carry over is everything you learned, along with the design. Treat the prototype's code as disposable and its findings as the asset.
Is a no-code MVP production ready?
It can be, for simple products at small scale, if the platform handles sign-in, data protection and backups properly. Check what the platform promises on each line of the checklist. Limits usually appear with custom logic, integrations, performance and cost as usage grows.
How much should a prototype cost compared with an MVP?
A small fraction. If a prototype approaches the cost of an MVP, it has stopped being a prototype. Its value comes from being cheap enough that you can afford to be wrong, and to try again.
Do investors expect a prototype or an MVP?
It depends on the stage. Early on, a prototype with evidence from real users can be enough to show the idea is worth funding. Later, investors generally look for a working product with customers. In both cases the evidence of demand matters more than the amount of software.
