No matter how good a concept paper may be, a client understands a digital solution better when they can see it, click on it and experience it for themselves. With AI, we can now bring that very moment forward significantly.
At Netresearch, we’ve been working in an agile way since 1998: showing something early on, gathering feedback, and refining it. What’s new is the speed. Nowadays, almost anyone can build a convincing clickable prototype in a short space of time. This changes when the key discussions take place in a project and what matters afterwards.
We’ve just experienced this first-hand with one of our own products, an AI tutor for workplace learning content.
Show it sooner, make better decisions
The first click-through prototype already mapped out the entire journey for a new customer: from the product page through registration to the finished course. Our managing director went through it himself, provided feedback, and all the changes were implemented that very same day – many within an hour.
This is more than just a gain in speed. Instead of discussing requirements on the basis of spreadsheets or concept papers, everyone is talking about the same experience. You can see where navigation isn’t working, where information is missing or where a workflow was conceived differently. Identifying issues becomes concrete and straightforward.
We now use this approach more frequently in client projects as well. We present information architectures early on as navigation rather than just as a table, and new features initially as wireframes or mock-ups. AI thus not only speeds up implementation but also facilitates understanding.
What looks finished is far from finished
Our clickable prototype subsequently became real software with an account, login and course upload functionality. The core functionality was in place within a single working day. At first glance, one might have thought: almost done. This is precisely where it gets interesting. Because nowadays, a compelling user interface says less and less about how mature a product actually is.
That’s why we don’t just build – we test systematically. To this end, we’ve developed our own AI-supported testing process and enriched it with technical and subject-matter expertise, as well as specific testing criteria. Among other things, it identified issues with access rights, data protection and unexpected user inputs, including a bug that would have prevented the system from launching. None of this was apparent during a normal click-through test.
And then there are the issues that don’t appear in any screenshot: encryption and certificates, retention periods, backups, logging, accessibility, operational security and legal requirements.
AI is also a huge help here. However, it does not replace the experience of knowing what needs to be tested, which risks are relevant and when a solution is actually ready for operation.
The value is shifting
This is a positive development for customers. They can view and evaluate things much earlier; misunderstandings become apparent more quickly and course corrections are less costly. At the same time, the value of professional software development is shifting. Building a good click-through mock-up is becoming increasingly straightforward. However, turning it into a secure, legally compliant and reliably operable product remains a challenging task.
Nowadays, the demo and the finished product can look almost identical. The difference lies in what you cannot see. And that is precisely why it is worth doing both at an earlier stage: making the idea visible and discussing what will turn it into a robust product.
No matter how good a concept paper may be, a client understands a digital solution better when they can see it, click on it and experience it for themselves. With AI, we can now bring that very moment forward significantly.
At Netresearch, we’ve been working in an agile way since 1998: showing something early on, gathering feedback, and refining it. What’s new is the speed. These days, almost anyone can build a convincing click-through mock-up in no time at all. This changes when the key discussions take place in a project and what matters most afterwards.
We’ve just experienced this first-hand with one of our own products, an AI tutor for workplace learning content.
Show it sooner, decide better
The first click-through prototype already mapped out the entire journey for a new customer: from the product page through registration to the finished course. Our managing director went through it himself, provided feedback, and all the changes were implemented that very same day – many within an hour.
This is more than just a gain in speed. Instead of discussing requirements on the basis of spreadsheets or concept papers, everyone is talking about the same experience. You can see where navigation isn’t working, where information is missing or where a workflow was intended to be different. Identifying issues becomes concrete and cost-effective.
We now use this approach more frequently in client projects as well. We present information architectures early on as navigation rather than just as a table, and new features initially as wireframes or mock-ups. AI thus not only speeds up implementation but also facilitates understanding.
What looks finished is far from finished
Our click-through prototype subsequently became real software with an account, login and course upload. The core functionality was in place within a single working day. At first glance, one might have thought: almost there.
This is exactly where it gets interesting. Because nowadays, a compelling user interface says less and less about how mature a product actually is.
That’s why we don’t just build – we systematically test. To this end, we’ve developed our own AI-supported testing process and enriched it with technical and subject-matter expertise, as well as specific testing criteria. Among other things, it identified issues with access rights, data protection and unexpected user inputs, including a bug that would have prevented the system from launching. None of this was apparent during a normal click-through.
And then there are the issues that don’t appear in any screenshot: encryption and certificates, retention periods, backups, logging, accessibility, operational security and legal requirements.
AI is also a huge help here. However, it does not replace the experience needed to know what needs to be tested, which risks are relevant and when a solution is actually ready for operation.
The value is shifting
For customers, this is a positive development. They can see and assess things much earlier; misunderstandings become apparent more quickly and course corrections are less costly.
At the same time, the value of professional software development is shifting. Building a good clickable prototype is becoming increasingly easy. Turning it into a secure, legally compliant and reliably operable product remains a challenging task.
Nowadays, the demo and the finished product can look almost identical. The difference lies in what you cannot see.
And that is precisely why it is worth doing both at an earlier stage: making the idea visible and discussing what will turn it into a robust product.
