Why Product Management Still Matters in the AI Era

Every few years a new technology arrives and someone declares product management dead. Mobile was going to do it. Then no-code. Then design thinking. Now AI. After 18+ years of building products, I have watched the tools change constantly and the job stay remarkably stable.

Here is the truth. AI is the biggest shift I have seen in my career. It compresses timelines, unlocks features that were impossible a year ago, and lowers the cost of trying things. But it does not tell you what to build, who to build it for, or why anyone should care. That is still the job.

The job has not changed

Our job: to delight customers in hard-to-copy, margin-enhancing ways.

I keep coming back to this sentence. It is the cleanest definition of product management I have found. Read it slowly. Delight customers. Hard to copy. Margin enhancing. If what you are building does not do all three, you do not have a product, you have a feature waiting to be commoditized.

The four pillars

As product managers, we own the full lifecycle. Todd Birzer frames it as four pillars that work together. This is the model I use with every team I advise.

Market Intelligence: Market, Customer, Competition

Strategy: Strategy development

New Product Development: Prioritization, Discovery and delivery, Launch

Lifecycle Management: Positioning, Pricing, Sales support, Sales channels, Product support, Finding growth, Obsolescence

Notice what is not on this list. Prompt engineering. Model selection. Vector databases. Those are tools. The pillars are the work.

What AI actually changes

Speed of research. What used to take a week of interviews and desk work can start in an afternoon. · Cost of prototyping. A working demo is now hours, not sprints. That changes how you validate. · Personalization at scale. Segments of one are finally realistic for teams that do not have Netflix budgets. · The bar for taste. When everyone can ship fast, the teams with strong opinions and clear judgment win.

What AI does not change

Customer empathy. AI can summarize interviews. It cannot feel the pain your user feels at 2am when your product breaks. · Judgment on what to build. A model will happily generate a hundred features. Picking the right one is still human work. · Prioritization. Roadmaps are trade-offs. Trade-offs need someone accountable. · Positioning and pricing. The market decides the value. Your job is to shape the story that makes it obvious. · Distribution. The best AI product with no channel is a science project.

Key takeaways

Use AI to compress the parts of the job that were slow. Do not let it replace the parts that require judgment. · Anchor every decision to the Biddle sentence. If it does not delight, is easy to copy, or shrinks margin, cut it. · Ship faster, but not sloppily. Speed without direction is just expensive noise. · Invest in taste and point of view. In an AI world, those are your moat. · Stay close to customers. The teams that lose in the AI era will be the ones that talked to their models more than their users.

The bottom line

AI does not make product management less important. It makes it more important, because the cost of building the wrong thing has never been lower and the cost of missing the right thing has never been higher. The tools change. The craft does not. If you want the practical version of that craft, read my guide to AI product discovery.

Framework credit: the four pillars diagram is adapted from Todd Birzer's Product Management 101, which quotes Gibson Biddle, former Chief Product Officer at Netflix.