The foundational principles guiding product development are undergoing a profound transformation, moving definitively away from a traditional project-centric methodology focused on delivering predetermined outputs, towards an agile, outcome-driven model centered on continuous discovery. This shift, long advocated by industry thought leaders, has been dramatically accelerated by advancements in technology, particularly the advent of generative artificial intelligence (AI), which has recalibrated the fundamental bottlenecks in the product lifecycle and redefined the critical contributions of product managers.
A Paradigm Shift: The Demise of the "Feature Factory"
For decades, and even persisting today, the most common approach to product creation has been the project model. This methodology is inherently focused on output. It typically involves senior stakeholders or executives conceiving and prioritizing a roadmap of features and projects. For each item on this roadmap, a designated "feature team product manager" meticulously crafts a detailed specification, often in the form of a Product Requirements Document (PRD). Subsequently, designers translate these specifications into visual designs, and engineers then build precisely to those designs. This sequential, hand-off heavy process, while offering a semblance of control and predictability, has frequently led to the creation of what industry observers critically term "feature factories" – systems highly efficient at producing features, but often disconnected from actual user needs or strategic business outcomes. The risk inherent in this model is that it can turbo-charge the production of ineffective or undesirable products, faster than ever before.
In stark contrast stands the product model, an approach fundamentally anchored in outcomes. Here, product leaders, often in collaboration with key stakeholders, first identify significant problems or opportunities within the market or user base. The responsibility then falls to an empowered, cross-functional product team. This team’s initial and crucial task is discovery: to iteratively explore and validate potential solutions that demonstrate a high probability of achieving the desired outcome. Only once robust evidence supports a solution’s viability and desirability does the team proceed to build and deliver it. A hallmark of this model is an intense focus on rapid experimentation and prototyping; it is not uncommon for product teams to generate 10-20, or even more, prototypes per week. This iterative, evidence-based approach to validation has been a cornerstone of effective product development long before the widespread adoption of generative AI, with tools like Figma already enabling high-velocity prototyping.
The New Bottleneck: Unlocking Worthwhile Solutions Through Discovery
The landscape of product development has evolved significantly, rendering the cost and speed of actual delivery far less prohibitive than in previous eras. The convergence of advancements in cloud computing, sophisticated DevOps practices, the rise of low-code/no-code platforms, and most recently, the transformative capabilities of generative AI in code generation, has dramatically reduced the time and resources required to build and deploy software. What once took months or even years can now be accomplished in weeks or days. This acceleration means that the primary bottleneck in product creation is no longer the engineering effort required to build a solution, but rather the intellectual and strategic effort involved in discovering a solution worth building.
A truly worthwhile solution is one that adeptly addresses both customer needs and the strategic objectives of the company. It must demonstrably generate the necessary business outcome – whether that’s increased revenue, improved user retention, or enhanced market share. Crucially, in a competitive market, a solution must not merely solve a problem, but solve it sufficiently better than existing alternatives to compel customers to switch or adopt it. This emphasis on superior value proposition places discovery at the forefront of strategic importance.
While AI undoubtedly assists in various stages of the product lifecycle, its contribution to discovery manifests in distinct ways compared to its role in delivery. In delivery, AI tools like Claude Code or Cursor can automate code generation, assist with debugging, and streamline testing processes, allowing engineers (and even technically proficient product managers) to accelerate the building phase. However, in discovery, AI serves more as an augmentation tool. It can analyze vast datasets to identify patterns, synthesize user research findings, generate diverse ideation prompts, and even simulate user interactions. Yet, the human element – the product manager’s deep domain knowledge, empathetic understanding of users, and strategic insight – remains absolutely critical for interpreting these AI-generated insights, formulating hypotheses, and guiding the iterative discovery process.
Redefining the Product Manager’s Imperative: "Build to Learn" vs. "Build to Earn"
As product teams navigate this transition from the project model to the product model, many product managers find themselves grappling with the evolving definition of their role and the nature of their contribution. The traditional responsibilities of project management and facilitation, while historically valuable, no longer provide the requisite strategic impact in an outcome-driven environment. This shift compels product managers to reconsider their core competencies and focus areas.
Some technically inclined product managers have recognized that with the advanced capabilities of modern engineering tools, including AI-powered coding assistants, they can directly engage in the development process, taking on more hands-on engineering tasks. While this path is valid for those with the technical foundation, it represents only one facet of the evolving role. The most astute product managers understand that while they are undeniably "product builders and creators," their building serves a fundamentally different purpose than that of engineers.
This distinction was eloquently captured by product coach Jeff Patton, author of User Story Mapping: Discover the Whole Story; Build the Right Product, who coined the phrase "build to learn vs. build to earn." This dichotomy powerfully articulates the differing objectives that drive activities in product discovery versus product delivery, a distinction that resonates with particular clarity in the current era of generative AI. Both phases involve "building," but the intent, tools, techniques, and even the definition of "testing" vary significantly.
Building to Learn: The Discovery Phase
In product discovery, the primary objective is to build to learn. This phase is a systematic endeavor to uncover a compelling combination of technology, functionality, user experience, and business constraints that effectively addresses the four key risks inherent in any new product or feature:
- Value Risk: Will customers use or buy this?
- Usability Risk: Can users figure out how to use it?
- Feasibility Risk: Can our engineers build it with the available time, skills, and technology?
- Viability Risk: Will this solution work for our business (e.g., sales, marketing, legal, finance, support, compliance)?
To mitigate these risks, product teams engage in rapid, iterative prototyping. With the current generation of tools, creating 10-20 prototypes or prototype iterations per week has become remarkably accessible, often without requiring direct assistance from product designers or engineers. These prototypes serve as tangible artifacts for testing. In the "build to learn" context, "testing" entails:
- Value and Usability: Engaging directly with target users and customers to gauge their interest, understanding, and willingness to adopt the proposed solution.
- Feasibility: Collaborating with engineers to assess the technical challenges and practicalities of implementation.
- Viability: Consulting with internal stakeholders across various departments to ensure alignment with broader business objectives and constraints.
Building to Earn: The Delivery Phase
Conversely, in product delivery, the mandate is to build to earn. This phase focuses on constructing a commercial-quality product that is ready for market, capable of being sold, serviced, and supported, and robust enough for customers to integrate into their operations or daily lives. The risks in this stage are markedly different, encompassing a comprehensive array of operational and technical considerations:
- Scale and Performance: Ensuring the product can handle anticipated user loads and perform efficiently.
- Fault Tolerance and Reliability: Designing for resilience against failures and consistent operation.
- Accuracy and Data Integrity: Guaranteeing precise functionality and reliable data handling.
- Privacy and Security: Implementing robust measures to protect user data and systems from threats.
- Operations and Maintainability: Ensuring ease of deployment, monitoring, and ongoing support.
- Provisioning and Internationalization: Supporting diverse deployment environments and global markets.
In the "build to earn" phase, "testing" takes on a rigorous, comprehensive meaning, involving quality assurance, performance testing, security audits, and compliance checks to ensure the product meets this extensive list of commercial demands and functions exactly as advertised.
Advancements in Discovery: Live-Data Prototypes and Parallel Exploration
The evolution of prototyping tools, especially those leveraging generative AI, has introduced two significant practical differences in contemporary product discovery.
Firstly, while the four major types of prototypes (sketches, wireframes, mockups, and functional prototypes) remain relevant and are utilized more than ever, the relative cost and speed of creating them have been fundamentally altered. The emergence of AI-powered tools has made the creation of live-data prototypes dramatically faster and cheaper. These are functional prototypes that can interact with real or simulated data, allowing product teams to place working versions of their product in front of select users and customers much earlier in the cycle. The ability to collect genuine usage data from these prototypes provides significantly richer and more authentic insights into user behavior and preferences, marking a game-changer for the "build to learn" philosophy. This capability allows for earlier validation of core assumptions and faster iteration based on actual user interactions rather than speculative feedback.
Secondly, the sheer speed and efficiency of AI-based prototyping tools now enable product teams to test multiple approaches in parallel. Traditionally, discovery often involved a largely sequential iteration process: a team would develop what they believed was the most promising solution, test it, refine it based on feedback, and continue this cycle until sufficient evidence emerged to proceed to productization. Only then would they "declare victory" and transition to delivery (building to earn). Today, however, it is increasingly common to rapidly generate several distinct prototypes, each exploring a different hypothesis or solution pathway to the same core problem. These diverse prototypes can then be tested simultaneously, allowing for parallel learning and a more comprehensive exploration of the solution space. This approach significantly accelerates the discovery process, reduces the risk of committing to a suboptimal path prematurely, and allows teams to converge on the most promising solution with greater confidence and speed.
This focused discussion on prototyping in discovery and productization in delivery underscores the core essence of the "build to learn" versus "build to earn" distinction. While other activities certainly occur in both phases, these building activities represent the fundamental drivers of progress and learning.
The Evolving Skillset for the Modern Product Leader
Reflecting these profound changes, many leading companies have begun to adapt their product management interview processes. The emphasis is increasingly placed on assessing a candidate’s understanding of this new paradigm, particularly their proficiency in building and testing prototypes as a core function of the role.
The skills and knowledge required for product managers to excel in a "build to learn" environment extend beyond mere technical aptitude. Becoming proficient with prototyping tools and mastering discovery techniques is, in many ways, the easier part of the journey. The more challenging, yet ultimately more critical, aspect is the development of robust "product sense." Product sense can be defined as an intuitive understanding of customer needs, market dynamics, and strategic business objectives, coupled with the ability to interpret diverse data points, evaluate learnings from experiments, and articulate a compelling vision for product direction. It is the discernment required to separate signal from noise in discovery data, to identify truly valuable solutions, and to guide the product’s evolution strategically.
It is important to acknowledge that not all product professionals are equally enthusiastic about embracing the "build to learn" philosophy or the hands-on nature of prototyping. Some prefer to define their contribution primarily as facilitators, managers, or as the "glue" that holds a product team together. While these coordination functions remain necessary to some extent, their strategic value as a primary differentiator for product managers is diminishing. For individuals who do not adapt to the builder/creator aspect of the modern product manager role, there is an increasing risk of their skills becoming outdated or marginalized.
However, for those who embrace the inherent builder/creator nature of the product manager role, actively focus on cultivating their product sense, and hone their "build to learn" skills, the current era presents an unprecedented opportunity. We are entering what many believe will be a golden era for skilled product professionals, where their ability to rapidly discover and validate valuable solutions will be paramount to organizational success and innovation. This demands a proactive shift in mindset and skill development, ensuring product managers are not just overseeing processes, but actively shaping the future through iterative creation and informed learning.
