In my four-year journey leading product teams across the technology sector, I’ve witnessed numerous shifts in how we approach product discovery. But none has been as transformative as what we’re experiencing with generative AI. This isn’t merely an incremental improvement to existing processes. It’s a fundamental reimagining of how we understand, anticipate, and fulfill customer needs.
Traditional product discovery has always been constrained by what customers can articulate. As Henry Ford allegedly quipped, “If I had asked people what they wanted, they would have said faster horses.” The challenge has never been collecting customer feedback; it’s deciphering the unspoken needs behind that feedback.
Generative AI is changing this paradigm. By analysing vast datasets spanning customer interactions, support tickets, social media, reviews, and usage patterns, these systems can identify patterns that human analysis would miss. They’re not just parsing what customers say, they’re interpreting what customers mean.
At Salesforce, the implementation of Einstein GPT allowed their product teams to analyse over 10 million customer interactions in 2023, revealing that 68 per cent of their B2B customers were struggling with a specific workflow that wasn’t captured in traditional NPS surveys. This insight led to the development of their Workflow Intelligence feature, which reduced customer task completion time by 37 per cent (Salesforce Annual AI Impact Report, 2024).
Product teams have historically operated reactively: we build, measure, learn, and iterate. This approach, while sound, means we’re always playing catch-up with evolving customer needs. Generative AI enables a more predictive approach.
Modern systems can now identify emerging needs before they become widespread pain points, simulate customer reactions to potential features without expensive prototyping, detect early signals of changing preferences or behaviours, and generate and test hypotheses at unprecedented scale and speed.
Spotify’s internal AI platform, Bandwagon, exemplifies this shift.
In early 2024, the platform analysed user engagement patterns and detected emerging interest in time-based playlists six months before this became a widespread user request. By the time competitors were reacting to explicit demand, Spotify had already shipped their “Daily Timeline” feature (Harvard Business Review, “How Spotify Uses AI to Anticipate User Needs,” March 2024).
Perhaps the most underappreciated aspect of generative AI in product discovery is its ability to synthesise insights across organisational boundaries. Customer data typically lives in fragmentary form across marketing, sales, support, product, and engineering teams.
Generative AI excels at connecting these dots. It can recognise that what marketing sees as a positioning challenge, support recognises as a usability issue, and engineering understands as a technical constraint, are all manifestations of the same underlying customer need.
This holistic view enables teams to address root causes rather than symptoms. At my current organisation, our generative AI system helped us identify that what we thought were three separate product initiatives were actually addressing different facets of the same customer journey. This insight allowed us to consolidate efforts and deliver a more cohesive solution three months ahead of schedule.
Despite these advances, I remain convinced that human judgment is irreplaceable in product discovery. Generative AI excels at pattern recognition, hypothesis generation, and data synthesis, but product managers bring critical elements to the process: ethical considerations, creative leaps, empathy, strategic alignment, and practical implementation knowledge.
The most effective product teams use generative AI as an amplifier for human insight, not a replacement. The technology helps us ask better questions, challenge our assumptions, and explore possibilities we might otherwise overlook. Looking to the future, I see product discovery evolving in several important directions: continuous discovery will become the norm, personalised product experiences will proliferate, cross-functional collaboration will deepen, product development velocity will increase, and product managers’ skills will shift toward hypothesis framing and strategic interpretation of AI-generated insights. The competitive advantage in the coming years won’t go to organisations with the most data or the most advanced AI, but to those who best integrate AI capabilities with human judgment to anticipate and fulfil customer needs.
Generative AI isn’t just another tool in the product manager’s toolkit, it’s a force multiplier that’s redefining what’s possible in product discovery. By helping us understand customers in unprecedented depth, anticipate their needs before they can articulate them, and test ideas with unprecedented efficiency, these technologies are enabling a more predictive, proactive approach to product development.
While the technology continues to evolve rapidly, one thing remains clear: the future of product discovery lies in the thoughtful integration of artificial and human intelligence. Those who master this integration will create products that don’t just meet customer expectations, they anticipate needs customers didn’t know they had.
Chika A. Nkwocha is a seasoned Product Lead with four years of extensive experience in the technology sector and a thought leader in the intersection of technology and product management, Nkwocha emphasizes the critical balance between artificial intelligence capabilities and human judgment. He advocates for an approach that uses AI as an amplifier of human insight rather than a replacement, highlighting the importance of empathy, ethical considerations, and strategic thinking in product innovation. Throughout his career, Nkwocha has been instrumental in helping organizations move from reactive to predictive product development models. He has a proven track record of using AI-driven insights to streamline product initiatives, reduce development cycles, and create more cohesive solutions that directly address underlying customer challenges.
