How PMs 10× Their Role with AI: Part 1

Welcome and Introductions

Rina Alexin | 00:00–03:00
Awesome. Well, we are going to get started. I’m sure more people are going to join. Excited to have this webinar today! We’re kicking off the first part of a three-part series on how Product Managers can 10× their role with AI. Today, Kenny and I are talking about the Product Manager bottleneck—where PMs have become the new constraint and how that shouldn’t be the case with AI.

Welcome, everyone. Please say hi in the chat and tell us where you’re joining from. We want this to be a conversation—so if you type once, you’ll find it easier to engage and ask questions later. Let’s kick it off. Kenny, over to you.

Kenny Kranseler | 03:00–04:30
Thanks, Rina. Hi, everyone! I’m Kenny Kranseler, calling in from Seattle, Washington. I’ve been at Productside for about six and a half years now, after a long run as a product manager at places like KFC, Microsoft, and Amazon—so I’ve got plenty of product scars to share!

About Productside and the Webinar Series

Rina Alexin | 04:30–06:30
For those of you unfamiliar with Productside—we’re an outcome-driven product partner. We help product teams transform how they work, combining training, advisory, and coaching, all tailored to context. Every engagement is led by experts like Kenny who embed with teams to design their future-state product function.

Before we dive in—yes, this session is being recorded. You’ll get the replay afterward. You can also join our LinkedIn group—over 60,000 product professionals—great place to share insights and connect.

This is part one of our three-part series:
– **Part 1 (Today):** Using AI in Discovery
– **Part 2:** Building Smarter with AI
– **Part 3:** Go-to-Market at AI Speed

Setting the Stage: The PM Bottleneck

Kenny Kranseler | 06:30–11:30
So, let’s start with the problem. There’s a growing crisis in Product Management—PMs have become the bottleneck in the AI era. Andrew Ng said it best: engineers are now 10× faster thanks to AI, but PMs haven’t sped up at the same rate.

Before we unpack that, quick poll: *How much faster are your engineering teams moving with AI tools?* 5–10× faster? 2–5×? Not faster at all? Or actually slower?

Looks like the responses are mixed! Some say no change, a few say slower, and only a small group reports 5–10× improvement. Interesting—and it sets up our discussion perfectly.

What Andrew got right: AI has made engineers incredibly fast at producing prototypes. What he missed is that the new bottleneck isn’t technical—it’s strategic. Deciding *what to build* is now the hard part. AI accelerates building; PMs must accelerate decision-making.

Why PM Workflows Lag Behind

Kenny Kranseler | 11:30–16:30
Engineering teams are upgrading their tools, but product workflows? Still lagging. Discovery is manual. Feedback is scattered across tools. Impact is hard to quantify. Prioritization is reactive and intuition-driven. It’s a systematic imbalance.

Andrew shared a striking example—some teams now talk about having *one PM for every half engineer*. That ratio would’ve sounded absurd a year ago. Yet it reflects how fast technical work has evolved—and how product hasn’t caught up.

But the answer isn’t “move faster.” Our job is to decide smarter. Doing everything in the backlog leads to bloatware, confusion, and wasted value. Just because we can move faster doesn’t mean we should.

AI Isn’t Replacing PMs—It’s Upgrading Them

Kenny Kranseler | 16:30–19:30
Here’s the good news—AI isn’t replacing Product Managers; it’s upgrading them. The best PMs of the future will be defined by clarity, speed of learning, and their ability to orchestrate impact.

Leading PMs are already redesigning workflows to use AI—not to automate judgment, but to create better, faster decisions. They’re reducing validation cycles from weeks to hours, synthesizing insights, and running discovery at machine speed.

Poll #2 – How Often Do You Use AI?

Rina Alexin | 19:30–20:30
Let’s take another poll: *How often are you using AI to enhance your product work?* More than daily? Weekly? Haven’t started?

Looks like three-quarters of you use it daily or more. That’s great!

Kenny Kranseler | 20:30–21:30
Exactly. AI won’t take your job. The PM who uses AI better will. Let’s talk about how that looks in practice.

Velocity Reimagined: From Speed to Impact

Kenny Kranseler | 21:30–26:30
Velocity isn’t about feature count—it’s about impact. As Zoom’s CPO Odag O’G put it, “Velocity means decisions and features that matter.” AI changes how fast we can move, but PMs must ensure that speed leads somewhere meaningful.

Discovery should no longer be a one-time phase—it’s a continuous experiment loop. AI gives us empathy at machine scale, helping us find patterns across thousands of data points instantly.

AI for Discovery: Empathy at Machine Scale

Kenny Kranseler | 26:30–30:30
AI allows PMs to process mountains of unstructured feedback—CRM logs, support tickets, and interviews—and surface actionable insights. It doesn’t replace your intuition; it sharpens it. Think of it as calibrating your gut with better data.

AI and Synthetic Users: Helpful or Harmful?

Rina Alexin | 30:30–33:00
Kenny, what do you think about PMs using AI agents to conduct user interviews?

Kenny Kranseler | 33:00–35:00
Synthetic users can be useful—if defined correctly. They can simulate real users up to 85–90% accuracy and help test ideas quickly. But they must be validated with real users. Use them to augment, not replace human feedback.

Rina Alexin | 35:00–36:30
Exactly. Don’t delegate your competitive advantage. Real empathy still comes from human connection.

Case Studies: AI in Discovery

Kenny Kranseler | 36:30–43:30
Companies like Dovetail, Maze, and WealthSmart have used AI to compress discovery cycles from weeks to hours. Dovetail automated feedback tagging, reducing 10 days of work to 3 hours. WealthSmart analyzed 500 interview transcripts in under an hour—previously a week-long task.

LaMier Insurance used AI to coordinate 1,500 specialists. They improved planning efficiency by 50% and reduced prep time from days to 20 minutes. The result? Faster alignment, better insights, and improved customer outcomes.

Vibe Coding: The Next Step (with Caution)

Rina Alexin | 43:30–46:00
We’re seeing a trend called “vibe coding”—rapidly prototyping software through natural language prompts. But let’s be careful. This is powerful, but dangerous if used too early in discovery. You risk skipping deep context.

Kenny Kranseler | 46:00–48:00
Right. Use it to visualize assumptions—not finalize solutions. Vibe coding bridges thinking and execution but shouldn’t replace exploratory research.

Audience Q&A – AI in Hardware and Emotion Capture

Rina Alexin | 48:00–53:00
Question from Derek: “Have you seen AI influence hardware prototyping?” Yes—Formula 1 and BP use AI simulations to test environments before building. It’s more about modeling than manufacturing, but hugely powerful.

Kenny Kranseler | 53:00–55:00
And to Najabat’s question about human emotion—AI can now detect frustration through facial recognition and sentiment analysis. It won’t replace ethnographic observation, but it’s another data signal to guide discovery.

Signal over Noise: From Feedback to Insight

Kenny Kranseler | 55:00–59:00
AI helps filter chaos into clarity. Nike used NLP to process thousands of reviews into real-time dashboards showing customer sentiment shifts. Instead of noise, PMs now get signal—weighted by customer value and churn risk.

Mini Validations and Continuous Learning

Kenny Kranseler | 59:00–63:00
The future isn’t about massive discovery projects—it’s about dozens of micro-validations. Small, cheap, directional experiments that compound confidence. PMs who adopt this rhythm evolve faster.

Recommended Tools for AI-Driven Discovery

Kenny Kranseler | 63:00–66:00
Here’s a quick preview of tools to explore—categorized for feedback clustering, hypothesis testing, and signal analysis. Treat them as accelerators, not decision-makers. AI shortens the distance between question and insight; you still frame the right questions.

Five Core Leaps for PMs

Kenny Kranseler | 66:00–70:00
1. **Outcomes over Outputs** – Focus on results, not releases.
2. **Synthesized Empathy** – Scale understanding, not just interviews.
3. **Data-Informed Instincts** – Combine gut + evidence.
4. **Continuous Learning** – Make discovery a habit, not a phase.
5. **Signal Focus** – Separate what matters from the noise.

Q&A and Closing Remarks

Rina Alexin | 70:00–75:00
Kenny, if someone’s just starting—what’s their first step?

Kenny Kranseler | 75:00–76:00
Simple: take your existing data—support tickets, CRM logs—and run them through AI tools. Let it summarize. You’ll be amazed how quickly insights appear.

Rina Alexin | 76:00–78:00
Perfect. And remember, vibe coding is last—not first! (Laughs) Thank you, everyone, for joining. Next up—Part 2: Building Smarter with AI, and Part 3: Go-to-Market at AI Speed.

We’ll also send out the **free AI Problem Starter Pack**—three prompts to help you clarify stakeholder asks and prioritize outcomes.

Kenny Kranseler | 78:00–79:00
Thanks, everyone. Great discussion. See you next time!