Product Management Is Dead, So What Are We Doing Instead? | Lenny & Friends Summit 2024
- Aisha Washington
- 2 hours ago
- 7 min read
Declaring product management “dead” is deliberately provocative. The argument presented at Lenny & Friends Summit 2024 is not that companies will suddenly stop making product decisions. It is that many routines associated with the product manager role—writing strategy documents, preparing slides, assembling feedback, producing wireframes, and coordinating handoffs—are being compressed or automated by AI.
That shift raises a more useful question than whether the job title will survive: what kind of product work will remain valuable? The speaker’s answer points toward AI-powered teams whose members operate across traditional boundaries, combine specialist judgment with generalist execution, and spend less time manufacturing artifacts around the work.
Product Strategy Must Look Beyond Today’s Workflow
The talk begins with a warning about the speed of technological change. AI has advanced quickly enough that assumptions about product design, engineering, and management can become outdated within months. Product organizations should therefore resist designing solely for current customer behavior or today’s team structure.
When setting direction, the speaker recommends considering what customers may need three, five, or even ten years from now. The same exercise should be applied internally: how might a product team operate in 18 months, three years, or five years? Which responsibilities will remain distinctly human, and which processes will be delegated to software?
This does not mean pretending to predict the future precisely. It means making explicit bets instead of treating the present operating model as permanent. Teams that examine those possibilities early will be better prepared to adapt when tools, customer expectations, and competitive conditions change.
AI Is Compressing the Strategy Cycle
Traditional product strategy can be painfully sequential. A product leader interviews customers and colleagues, gathers input from sales and engineering, organizes the resulting notes, and spends days or weeks drafting a substantial document. That document then circulates through rounds of comments and revision.
Drawing on a past experience as a chief product officer, the speaker describes how this familiar process can expand into a long, iterative undertaking. Much of the effort goes into synthesizing information and polishing the artifact rather than testing whether the underlying choices are correct.
Generative AI dramatically shortens that production cycle. A tool can turn source material into a credible initial strategy draft within minutes. The output is unlikely to be ready for approval, but it may provide a useful 75% or 80% foundation. A product leader can then devote attention to the difficult parts: correcting assumptions, sharpening tradeoffs, applying context, and committing to a direction.
The important change is not that AI can write ten pages. It is that producing those pages no longer needs to consume the scarce time of an experienced product leader. Judgment remains central; document construction becomes cheaper.
The Product Manager’s Artifact Factory Is Disappearing
Strategy documents are only one example. The speaker identifies several other areas in which AI and no-code tools are reducing the labor required to move from an idea to something others can evaluate.
Wireframes once progressed from paper sketches to specialized design software. Product managers can now generate interactive prototypes quickly enough to put a working experience in front of colleagues or customers during the earliest stages of discovery. Feedback analysis can likewise move beyond manually sorting spreadsheet rows. Automated workflows can collect, classify, summarize, and prioritize recurring signals. Presentation tools can draft slides instead of forcing teams to construct every deck from a blank page.
Common coordination work is also becoming easier to automate:
Drafting routine documents and status updates
Summarizing meetings and decisions
Organizing feature requests and customer feedback
Monitoring goals and OKRs
Tracking competitors and market changes
Preparing an initial analysis for review
The speaker’s practical standard is speed to a useful intermediate result, not perfect automation. If a system can complete three-quarters of a task reliably, a person can use expertise to finish the remaining quarter. Accumulated across a week, those savings can return days to a product leader.
Becoming an AI-Powered Product Team
Automation is only the first part of the proposed response. The larger opportunity is to reinvest saved time in capabilities that expand what product people can do.
The speaker encourages product managers to learn adjacent skills such as rapid prototyping, design, coding, data analysis, and commercial thinking. The goal is not to become equally accomplished in every discipline. It is to develop enough practical range to remove bottlenecks, test an idea directly, and advance a project without waiting for every traditional handoff.
A product leader named Cody is offered as an example of this broader profile. Although he does not fit the conventional product-manager template, he learned design and coding skills that helped him navigate obstacles and deliver results. His value came from being able to cross functional boundaries while still applying product judgment.
This kind of development should not depend on a few unusually motivated individuals. The speaker argues that leaders must help entire teams automate routine work and acquire complementary skills. Internal channels can make that learning visible by giving people a place to share workflows, demonstrate experiments, and request help. Organizations also need to fund the necessary tools and normalize their use; encouragement without access or time will produce little change.
From Functional Handoffs to Generalist-Specialists
The familiar product trio—product manager, designer, and engineer—is often celebrated as a model of collaboration. Yet the speaker argues that it can become a disguised handoff system. The product manager specifies, the designer visualizes, and the engineer builds. Each participant retains a lane, but no one necessarily owns the full path from problem to outcome.
The proposed alternative is a generalist-specialist model. People still possess deep expertise in product, design, or engineering, but they participate more broadly in whatever work is needed to move the project forward. The principle is simple: there are no fixed lanes.
That does not make specialist knowledge irrelevant. Strong engineering, design, and product judgment remain essential. What changes is the assumption that only one function may perform a particular activity. A designer might build a prototype, an engineer might shape positioning, and a product specialist might analyze data or implement a small experiment.
AI makes this overlap more practical by lowering the technical cost of unfamiliar work. It allows specialists to reach farther without requiring every task to enter another team’s queue.
The AI-Powered “Triple Threat”
Taken further, this model produces what the speaker calls an AI-powered triple threat: a person capable of leading across product, design, and engineering while coordinating AI tools, agents, and platforms.
Such a person is not simply doing three full-time jobs. Instead, AI collapses parts of the talent stack by making execution in adjacent domains more accessible. A capable individual or very small team can prototype, analyze, design, build, and communicate with far less operational overhead than before.
The resulting environment favors opinionated product leadership. When production becomes faster, indecision can become a larger constraint than implementation. Leaders must understand customers, choose a direction, connect product work to commercial outcomes, and use evidence without hiding behind endless process.
The talk acknowledges that this prospect can feel threatening. Broader roles may mean fewer conventional product-management positions. Yet the speaker frames preparation as more constructive than denial: people can build range now, while organizations can redesign responsibilities deliberately rather than waiting for change to force a rushed reorganization.
Product Leadership Will Change Too
AI does not affect only individual contributors. Product executives will need to manage unfamiliar combinations of people, software agents, and external tools. Their budgets may increasingly involve tradeoffs between hiring additional employees and purchasing systems that increase the capacity of an existing team.
Leaders will also need enough technical fluency to manage people who code and enough commercial awareness to connect product decisions with distribution, marketing, and revenue. Team design will become less standardized. Rather than copying a universal product trio across the organization, leaders may assemble each team around the particular strengths and constraints of its mission.
The speaker describes this as a more handcrafted approach to organizational design. One initiative might revolve around a person with exceptional product insight; another might require strong engineering depth or unusual market expertise. Leadership’s task is to identify the scarce capability and construct the surrounding team accordingly.
Culture matters as much as structure. If people are punished for crossing functional boundaries or experimenting with new tools, the organization will preserve old bottlenecks even after acquiring AI capabilities.
Build Skills That AI Cannot Easily Replicate
As routine production becomes cheaper, durable advantage moves toward judgment formed through experience. Strategic thinking, taste, customer understanding, organizational influence, and the ability to make decisions under uncertainty are harder to reproduce than a status report or first-draft roadmap.
The speaker therefore urges product professionals to develop a distinctive value proposition. That could mean learning to build and scale AI-powered teams, gaining real technical competence, strengthening commercial instincts, or combining deep domain knowledge with fast execution.
The aim is not to compete with AI at tasks it can perform instantly. It is to use AI as leverage while becoming better at deciding what deserves to be built, recognizing when an answer is wrong, and aligning people around consequential choices.
Reanimating Product Management
The talk’s final argument is more optimistic than its title suggests. Product management can be “reanimated” by abandoning processes that no longer justify their cost. When AI handles more drafting, synthesis, prototyping, and coordination, product professionals can spend more time with users and apply more creativity to the problems that matter.
Achieving that outcome requires action at both the individual and organizational levels. Product people need to automate routine tasks, broaden their abilities, and teach colleagues what works. Leaders need to provide tools, budgets, psychological safety, and authority to people who can operate across domains.
The future presented here is not free of disruption. Roles may merge, teams may shrink, and established career paths may become less predictable. But the essential work—understanding needs, making strategic choices, and turning ideas into valuable products—remains. The opportunity is to perform that work with less ceremony and greater range.
