Leveraging AI and Data Science for Early-Stage Product Strategy

In an age where a well-defined product strategy can make or break a startup’s success, leveraging advanced technologies like AI and data science is becoming essential for organisations, and expected for tech-driven companies. To explore how these tools can shape effective product strategies, we sat down with Dr. Olumide Okubadejo, PhD, Head of Product at Sabi and a member of the Ventures Platform Expert Network.

With extensive experience in product innovation and machine learning, Dr. Okubadejo shares practical insights into using AI and data science to refine customer personas, drive scalable growth, and anticipate market shifts. In this Q&A, he provides actionable strategies for founders aiming to develop adaptable, data-driven products that meet evolving customer needs and maintain a competitive edge.

‍

1. When refining customer personas, what strategies do you recommend to ensure that the product consistently addresses evolving market needs?

‍

Kevin: The following strategies come highly recommended to ensure that your product consistently addresses your evolving market needs.

Treat personas as evolving profiles and update them based on new insights and data: Market needs are even more dynamic these days. To adapt to market dynamics, customer personas must be dynamic and proactive. It is incumbent on us to treat personas as evolving profiles rather than static documents. We should regularly revisit and update them based on new insights and data. To achieve this, I have come to rely on several data-oriented methodologies. 

Create customer feedback loop for action-oriented data insights: Establishing regular channels for customer feedback through surveys, interviews, and user testing is essential. This engagement is as though one is in constant dialogue with the user. This ongoing dialogue helps you stay attuned to customer needs and preferences as they change over time. The end goal of data is action. However, to get there, one has to go through insight. Leverage analytics tools to monitor user behaviour, engagement metrics, and purchasing patterns. Analysing this data can reveal trends and shifts in customer needs that inform persona updates.

Experiment for adaptability and strong customer focus: Experimentation is also vital to learn about your customers and update the personas. You should experiment with product features or marketing messages to see which resonates most with your audience and adjust personas accordingly. These strategies allow early-stage startups to remain adaptable and maintain a strong customer focus even as they scale.

‍

2. How can startups balance rapid scaling with maintaining a high standard of product quality and user experience?

Kevin: This is quite a complex thing to optimise for. On the one hand, scaling rapidly, being quick to the market, testing ideas quickly, and iterating effectively is necessary. On the other hand, it is expedient to build right. I have mostly balanced both by having ways to test out ideas, rapid prototyping, and RnD before investing significant engineering resources in a problem. Once an idea/feature is tested and validated through a prototype build, I employ specific industry techniques to push its engineering boundaries.  Most of these techniques involve adopting agile methodologies for iterative development, quick feedback incorporation, and rapid response to issues.

One question that arises is how features are prioritised for development or experimentation. Many product teams use the MosCow framework, but I particularly like the KANO framework and encourage my team to use it. Other points to note are the investments in scalable engineering infrastructure and automated testing tools.

On the issue of user experience, keep it central by involving UX designers throughout the development process and conducting regular usability testing.

‍

3. When leveraging data science and machine learning in product development, how can early- stage startups use these technologies to better understand and refine their product-market fit?

Kevin: As startups progress through their journey to unicorn status, they can leverage data science and machine learning to enhance their products in numerous ways. By analysing historical data, machine learning can empower startups to anticipate future customer needs and enable them to make informed decisions about product features, pricing, and marketing strategies. Additionally, startups can utilise algorithms to segment customers based on shared behavioural patterns and preferences. This allows product teams to build tailored product offerings, understand churn, and allow targeted marketing efforts. 

I always tell my team that the ultimate goal of search and recommendation is to understand the user's mind. Machine learning can analyse user data to deliver personalised experiences, such as recommending products or content based on individual preferences. I have tried not to include the obvious use case of personalised chat.

‍

4. AI and machine learning are powerful tools for predicting trends and user preferences. How can startups integrate these technologies to anticipate customer needs and stay ahead of market changes while scaling?

Kevin: Well, they can implement AI systems that analyse user data in real time to identify emerging trends and shifts in behaviour. Building these tools into your software/product before you deploy is non-negotiable. The ability to collect data and analyse is critical. I worked at a company that would collect all sorts of data. If they realise that users are clicking a button twice or multiple times, they automatically know that the UI is broken.

AI can also be deployed to monitor industry news, competitor actions, and market indicators, providing insights that inform strategic decisions. The user journey can be analysed to identify friction points and opportunities for enhancement, improving overall user experience. 

‍

5. Given your experience teaching AI and business intelligence, how do you see emerging technologies like machine learning shaping the future of product strategies in high-growth companies? What should startups consider when incorporating AI into their product roadmaps?

Kevin: Machine learning and AI will significantly impact product strategies in high-growth companies. They will enable data-driven decisions by analysing vast amounts of data, leading to more informed decision-making. AI will also drive personalization, allowing for highly customized user experiences. Routine tasks can be automated, freeing up resources to focus on innovation and strategic initiatives. However, I have never been a fan of incorporating AI for the sake of it. I always posit that incorporating AI should come from three angles, namely in order of importance;

‍

1. Understanding user behaviour 
2. Solving operational challenges (back office)
3. Solving user-facing problems.

‍

I have spoken in detail about user behaviour and would only touch on 2 and 3. I am strongly aligned with the strategy that AI is more effective when it is used to deliver better value rather than being the value in itself. As such, when I think and deploy ML technology, I always seek to use it to solve operational challenges such as efficient Marketing, PR, etc.

‍

Startups incorporating AI into their product roadmaps should consider several factors. Strategic alignment ensures that AI initiatives align with the company's overall goals and product vision. 

‍

______

‍

‍

Dr. Olumide Okubadejo, PhD, has a strong background in image processing, data science, computer vision, and machine learning, with a PhD in AI Image Processing from Université Grenoble Alpes, and a Master's Degree in Artificial Intelligence from University of Southampton. He is also a guest lecturer at ESIEE PARIS, where he teaches data science, business intelligence and artificial intelligence to masters students. He is also a certified instructor on LinkedIn, where he has completed multiple courses on topics such as design thinking, leadership, and communication.

‍