In a recent LinkedIn post, product discovery expert Teresa Torres offers a measured perspective on the rapid advancements in Artificial Intelligence, emphasizing that many core product discovery principles remain unchanged and are perhaps more critical than ever. Amidst the flurry of “hot takes” about AI’s transformative power, Torres urges leaders to ground their strategies in foundational practices.
Torres begins by directly addressing the prevailing narrative, stating:
“I keep seeing all these hot takes on what has changed with AI. I’m going to share what hasn’t changed.”
This sets the stage for her argument that while new technologies offer new possibilities, the fundamental approach to understanding customer needs and building successful products relies on enduring habits.
Starting with Outcomes Remains Paramount
A key point Torres highlights is the continued importance of defining clear outcomes before diving into development. As she notes, knowing what you are trying to achieve is still the bedrock of effective product strategy.
“If you don’t know what you are trying to achieve, it’s still very hard to achieve it.”
This principle, according to Torres, is crucial regardless of the technological tools available. Without a clear objective, even sophisticated AI tools cannot guarantee success.
Customer Interviews: A Timeless Practice
Torres also defends the enduring value of customer interviews, particularly story-based interviewing. She argues that this method is not about discussing current products but about understanding the customer’s world, their past experiences, and their goals.
According to Torres, even with AI, these interviews are invaluable:
“Yes, AI makes a lot just now possible. But story-based interviewing was never about asking about your product. It’s about asking about your customer and what they are trying to accomplish. And then collecting specific instances about how they do that today. This is not new. This is helpful even with new technology because it exposes the needs, pain points, and desires that can now be addressed in new ways with the technology.”
In her view, understanding these core needs is essential for leveraging new technologies effectively.
Metrics and Assumptions: The Foundation of Evaluation
Torres also touches upon the nature of evaluations, asserting that they are fundamentally metrics. She warns against focusing solely on technological performance without considering user needs.
As she puts it:
“With any metric, if we only focus on technology performance and not user needs and whether or not we are meeting those needs, our metrics are measuring the wrong thing. Same with evals.”
Furthermore, Torres cautions against a purely experimental approach without testing underlying assumptions. She argues that simply building something and observing the results is insufficient if one doesn’t know what to fix when it fails, likening it to guessing without a systematic process.
Addressing Delivery and Feature Bloat
While acknowledging that AI might alter delivery bottlenecks, Torres remains skeptical about whether delivery is truly no longer the primary constraint. She points out that customers have limited bandwidth for new features, reinforcing the timeless challenge of feature factories and feature bloat.
In conclusion, Teresa Torres encourages professionals to embrace new AI technologies while remaining anchored to fundamental discovery habits. She advises:
“Yes, learn the new technology, be mindful of what is just now possible. And keep working the discovery habits. Don’t let all the nonsense out there convince you to fall back into bad habits.”
Her message is a call to integrate AI capabilities with established, research-backed discovery practices, ensuring that innovation serves genuine user needs and business outcomes.
📝 About This Content
This article is based on insights shared by Teresa Torres on LinkedIn.
📅 Originally posted on September 11, 2026 | View original post on LinkedIn →