top of page

Validation first. AI as just another digital project

When we feel urgency about something, we automatically react instinctively. Whether the urgency is positive (enthusiasm) or negative (fear of missing out), the end result is the same: emotions, not rationality, prevail, even in otherwise rational people. The hype around AI has created a similar urgency in the SME sector. Many players who have been involved in digital product development are now focusing on AI, but it is worth thinking through a few things before you put all your eggs in one basket.

Read Bobcats Coding's expert insights on why SMEs should prioritize leveraging pre-existing AI technologies or LLM over attempting to build from scratch.

We have seen similar enthusiasm for digital developments in recent decades. And if we haven't learned the basics in the almost quarter century since the 2000s, AI will teach us - and it will be a costly lesson. Two lessons from digital product development that seem to be left behind these days, that have already proven to be verified: 

  1. Make sure that the product you are developing meets a real customer need. 

  2. Choose the right methodology for validation.

About the real needs

SMEs are unlikely to be large enough to affect the whole market and create new demands (most large companies are not big enough to do this). But they are close enough to their own customers to recognize their existing needs. The metrics to be examined include:

  • What need does the product satisfy / what fear does it address?

  • Who is the target audience? How many of them are there? 

  • In what way are these needs being met now?

  • What is the purchasing power of the target group? (How much are they spending now to satisfy the need? How often do they purchase? When was the last time they spent?)

  • What data sources do you have / what data sources can you create to make data-based decisions? Are they enough to train an AI model?

  • How long is my build-measure-learn cycle and how can it be shortened?

There may be more of these questions depending on the product, but one thing is for sure - there is no room for vanity or wishful thinking in this step of development. Of course, if you need a trusted strategic partner, you can always contact our experts.

About the appropriate methodologies

Once you know what you want to find out, it is important to choose the right methodology. These are typically set up to deal with various market uncertainties, and there is a wide range to choose from. (Still, forget about traditional waterfall. Do not start development in a vacuum, relying only on our own ideas, investing all the resources into the product that only meets the market after it’s finished.)

  • Agile methodology. Use this, if you already have some information about your future clients. Scrum, Kanban, and Extreme Programming - you know the drill. 

  • Lean startup and MVP. A rigorous methodology, based on the scientific method for research, in case you don’t even know who your customers might be: rapid experimentation, validated learning, and iterative development from the very beginning. 

  • Design thinking. A human-centered approach to innovation and problem-solving that focuses on understanding user needs, ideating creative solutions, prototyping, and testing. 

  • DevOps. These practices enable digital product teams to achieve continuous integration, continuous delivery, and faster time-to-market by automating build, test, deployment, and monitoring processes.

  • Rapid Application Development (RAD). These techniques include using visual modeling tools, reusable components, and user feedback to accelerate the development process and deliver working software within short timeframes. 

  • User-Centered Design (UCD). A process that is focused on understanding user behaviors, preferences, and goals to create digital products that are intuitive, usable, and satisfying to use. It involves user research, usability testing, and iterative prototyping

These digital product development methodologies can be tailored and combined based on the specific requirements, goals, and constraints of each project, enabling teams to deliver high-quality digital products that meet user needs and business objectives effectively. The development of an AI product is much more expensive than building a traditional digital product, so it is important to make sure that all our efforts are going in the right direction. 

We believe SMEs can and should join the AI race, but they should also prepare for a marathon. Consider how you can prepare before the big challenge by remembering what you have already learnt about digital product development - and put it into use.

AI Innovation Guidebook with LLM solutions

30 views0 comments


bottom of page