Proof of Concept 

A Proof of Concept (PoC) is a quick, technical test designed to determine whether an idea is feasible. When there’s doubt about whether a concept can work from a technical perspective, a PoC helps you understand the potential technical challenges. However, a PoC is only necessary if there’s a clear technical uncertainty that needs to be addressed.

Together we clearly define the steps that users and systems must take to bring the idea to life. The PoC will focus on the specific part of the concept where there is a feasibility risk—only testing what’s critical to making the idea work and where there is no evidence yet that it can work.

The timeline for a Proof of Concept (PoC) varies depending on the complexity of the project but is designed to be completed quickly. A PoC is conducted early in a project when a technological challenge arises, and you need to test whether a specific solution or technology can address the issue. Instead of fully implementing the solution, the PoC focuses on validating its feasibility, providing enough information for a go/no-go decision without refining all the details.

The goal of a PoC is to minimize effort while proving that the core concept can work. Ideally, it should take only a few days to a few weeks, depending on how complicated the problem is. This rapid testing helps the team decide whether to move forward, adjust the approach, or abandon the idea altogether, balancing speed with the need for meaningful insights.

Example 1: One recent PoC involved a client in the travel industry who needed to combine data from multiple APIs to create a niche solution. The challenge was that they didn’t have software engineering skills to integrate these APIs, and it was unclear whether the data would meet their use case. To help, we developed a small PoC—a thin software wrapper around the APIs—that allowed their team to experiment with the data and validate their idea, without committing to expensive contracts with external agencies. This gave them quick insights on feasibility before taking further steps.

Example 2: In another example, we've worked with clients exploring the use of generative AI (GenAI), like ChatGPT, for specific tasks. While GenAI models are powerful, their performance for a given use case can’t be fully assessed without testing. For some projects, experimenting in web interfaces like OpenAI or Gemini was sufficient, but others required building technical infrastructure. By running small PoCs (usually in just 3-5 days), we helped teams better understand the cost, potential pitfalls, and value of integrating AI models into their workflows.

Example 3: We’ve also worked with clients on medical applications and chatbot solutions designed to assist with clinical decision-making and diagnosis. However it was unclear how this technology could positively impact the life of a user. There was no clear opportunity defined. We emphasized the need for a clear opportunity and measurable outcomes before diving into technical solutions. This way the PoC could be clearly evaluated.

Example 4: To streamline customer onboarding, a client wanted to use a chatbot, ensuring users got up to speed quickly while maintaining consistent quality. We started with a clear desired outcome and, to explore all options, we encouraged the team to develop at least two alternative solutions to achieve the same goal. Using a user story mapping technique, we outlined these ideas, which provided a solid foundation for identifying key assumptions—factors that needed to be true for the idea to succeed. Some of these assumptions were technical, while others related to value, viability, usability, and ethical considerations. We identified the critical technical assumptions that posed the most uncertainty and conducted a focused PoC to test their feasibility. This allowed the client to validate the most important aspects of the solution, ensuring that the chatbot could deliver on its promise while minimizing risk.

We tailor PoCs by first ensuring we understand the specific technology you need and closely collaborating with your team. Together, we define the scope to ensure the PoC addresses your most pressing technical challenges while aligning with your business objectives and industry challenges. We ensure that the PoC focuses on the areas of greatest uncertainty. This ensures that the technical insights gained from the PoC are directly relevant to solving real-world challenges in your industry, whether it’s healthcare, finance, retail, or any other sector.

Using a Proof of Concept (PoC) can be an invaluable step in your project development. The real value lies in the experiences and insights you gain while creating the PoC. While you might think of the PoC as a tool for demonstration, its primary purpose is to help you test the feasibility of your idea before committing significant resources to a full-scale implementation.

It's important to distinguish between a quick PoC built to gather feedback and a fully developed product ready for customers. A polished product requires more time and effort, including a user-friendly interface, additional functionality, robust cybersecurity measures, and quality assurance tests. While these elements are crucial for a product intended for actual use, they can slow down the feedback process.

The goal of a PoC is not to serve as a sales tool but to minimize technical feasibility risks. By addressing these risks, you can build the confidence needed to invest further in creating a product for the market. Additionally, consider other types of risks:

• Value Risk: Will customers prefer your product over alternatives?

• Viability Risk: Can your business effectively sell and market this idea? Is it compliant with regulations? Can you generate profit?

• Usability Risk: Can users easily find and use your product?

By addressing these factors, you set a solid foundation for future development, ensuring that your final product meets market needs and stands out against the competition.

A Proof of Concept through sustAIn.brussels is free due to the access to de minimis which is accessible for SMEs, Startups and Midcaps.

SustAin.Brussels draws from the combined expertise of our consortium members. Sirris, the collective centre of the Belgian technological industry is a reference in technology adoption. They have skilled software engineers with extensive hands-on experience and data scientists who leverage data effectively to drive insights.

Through our consortium members ULB & VUB we have access to over 40 years of combined experience in artificial intelligence.

The biggest risk with a Proof of Concept is trying to do too much with it. A PoC should stay focused on testing the core idea or technology, but sometimes there's a temptation to keep adding more features and functionality. This can lead to spending too much time and resources on a PoC instead of moving into a proper engineering process that follows best practices for long-term development.

Aside from that, a PoC is generally low-risk and can actually help you reduce overall project risks. It provides fast feedback, allowing you to make better, informed decisions about whether or not to move forward with the technology.