I’ve always felt that prototyping is the best way to design & spec a solution. It’s an approach that’s been around for 30+ years and the benefits are significant.

For the uninitiated, prototyping a solution is the act of building a working model before committing to the full build.

Getting a demonstrable working version that really looks and feels like the finished product in front of key stakeholders like users, customers, or the board is so much more engaging than slides, wireframes, documents or any other surrogate attempting to represent a project’s intended output.

Why prototyping matters

Giving users something that actually works, allows them to consider how well the proposed solution might support their daily tasks in a way that a wireframe just doesn’t. It’s impossible to overstate how valuable that extra level of engagement is in flushing out issues early while it’s still cheap and easy to accommodate change.

It also eliminates the feeling of fear that can build up in a team that knows a new system is coming but haven’t seen it or have failed to grasp the intent from documentation that they have seen. Instead they feel confident and keen for the arrival of a system they have had a fair hand in designing.

Similarly when trying to share vision with potential backers or senior management, a fully working model ensures that everyone is on the same page and allows people to not just grasp ideas but actually build on them too. All of which increases engagement and the chances of success and approval.

The problem before AI

The problem before the advent of AI was that producing a proper working prototype wasn’t much less effort than building the real thing and simply far too expensive for pre-project support. When prototypes were created in the development process, by the time a satisfactory working model had been created, nobody was prepared to throw it away and start over on the real build. Instead the prototype just continued to be developed and ultimately pressed into service, missing the point of the exercise and resulting in a less than ideal outcome!

AI changes everything

However, now that we have AI we can generate prototypes in no time at all and for negligible cost, bringing an idea to life while the idea is fresh, allowing momentum to develop. It’s then easy to evolve that prototype with maximum user engagement to a point of perfect alignment with your requirements, where all parties are satisfied before pushing the button on full development.

The ability to work quickly and effectively means we can now think about application design in a much more pure way, really focusing on the user experience (UX), deploying our expertise and imagination to create solutions that don't just work, they make everyone happy.

A real-world scenario

Just imagine, you have an idea or a requirement but you’re not sure how it will be in practice. You commission a prototype and a couple of days later you’re looking at your solution and working through a few use cases.

The prototype runs in your browser using HTML & JavaScript and with an IndexedDB database, which means you have a fully functional application that feels incredibly close to the real thing. Most users probably won’t even realise it’s not a full production app!

You can see a few flaws so you evolve it quickly to V2 before sharing it with the team. They get what you are thinking and make a few helpful suggestions, which go straight into the next iteration. You take it to the board for approval and they love it too and are reassured by the level of buy-in from the team. Approval follows swiftly and you’re off to the races!

The prototype remains as the development project’s north star, but the coding can start afresh because you haven’t spent a fortune building the prototype.

If you’re building with Provanta then you will also be pleased that the prototype is not only used as a reference for the development project, it is actually used as part of the spec that is passed to the AI in generating the code of the final system, speeding up the process and ensuring maximum compliance.

The bottom line

Prototyping is not a new methodology but having been enabled by AI it’s a huge benefit in the development process.

If you want to explore more or have us build a prototype for you, please drop me a line!