If you only read one section this week:

• B2B CMO or PMM? The first section about product marketing and AI.

• Everyone else: if you use Google NotebookLM or Workspace, their ongoing improvements.

-François

5 Ways Top Product Marketers Use AI

I have interviewed enough Product Marketers recently (esp. in AI, DevTools, and Cyber) to know that the gap between "good" and "great" PMM also shows in how they use AI.

Most candidates tell me they use ChatGPT to do some research, write messaging, launch blog posts, craft whitepapers, etc.

That is now table stakes.

The best PMMs also use AI as a thinking partner, advanced analyst, persona simulator, asset grader, and enable their entire team + their AI with deep context.

Here are 5 of their best AI practices:

1. Accelerating their own onboarding

You can’t be a great PMM without first building up your product and market expertise - including how and why customers buy your products or your competitors.

The best PMMs use a method and a cadence to learn what they need (e.g., "I review X Gong calls every Friday").

To ramp up on a new technical product, they load documentation, PRDs, Sales Engineer decks and meetings, and architecture diagrams into AI tools like NotebookLM or Projects in ChatGPT and Claude.

Then, they create learning artifacts with AI (e.g. flashcards, podcasts) and get quizzed (NotebookLM has that built in). They chat with it until they can explain the technical nuances back to the AI.

2. Building the "Central Marketing Brain"

AI-first PMMs use AI to synthesize large amounts of intel into a single source of truth that centralizes essential context for their colleagues and their own AI chats.

a. Detailed ICP and persona profiles: for each buyer and user segment, they feed sales call recordings, first- and third-party research, deep internet research, and online reviews, into NotebookLM or Projects, to build full reference profiles.

b. Intel about product, market and competitors. They also add to the brain-approved product messaging, market and competitive intel, case studies, objection handling, core themes for thought leadership, analyst reports, etc.

Colleagues in Marketing, Sales, and Product then use this "brain" as context for their own AI work. This ensures everyone operates from a shared deep understanding of the product, buyers, and market.

3. The Simulator and Grader Loop

Once the "brain" is built, these PMMs use these knowledge bases to pressure-test and improve everything: pitches, content, launch plans, etc.

They configure a Custom GPT or Project as Synthetic Personas or asset graders to role-play a specific target profile (e.g., a skeptical VP Eng or an alert-fatigued security analyst). They show it a draft pitch or blog post and ask for brutal, structured feedback and specific improvements.

They ask:

• "Is this technically accurate for product XYZ?" "Would a senior dev roll their eyes at this sentence?"

• "Tell me which section will raise objections, why and how to adjust."

• “What would you change to drive them to action?”

They use AI to kill buzzwords and iterate until the AI "buyer" gets it and stops objecting.

4. Voice of Customer (VoC), at scale

PMMs cannot read and synthesize every support ticket, Reddit thread, online review, or Discord message. But AI can. Every day, if needed.

AI-savvy PMMs set up systems to find and ingest unstructured data from multiple sources. They use AI to cluster the feedback into themes and recommend product improvements, messaging tweaks, new assets, or campaigns.

They quickly find non-obvious insights: e.g., a missing feature, a confusion onboarding step, or repeat complaints about competitors’ products that they can exploit.

Their AI even generates feature recommendations with user stories ot campaign briefs on the fly.

5. A Sales Enablement coach

Static PDF battle cards stored in an LMS are where good intel goes to die.

The best PMMs have moved to dynamic enablement. They build NotebookLMs, custom GPTs, or Projects that Sales reps can chat with or get quizzed by (as per #1).

Instead of reading a document, a rep can ask the AI: "I have a call with a CTO in the finance sector who is worried about compliance. How do I handle that question? What are our relevant case studies or customer quotes? Coach me through a mock conversation."

This turns enablement into an active, 24/7 coach vs a “one-time and forgotten” thing.

These are just 5 of their best practices. AI can also be tapped to create pitches, landing pages, and assets for micro-segments, to build segment-specific ROI Calculators, and much more.

The possibilities seem endless.

🛠️ My selection of resources, tips, and tools

📈 A marketing dashboard for the AI era

From Kieran Flanagan (yes, again), SVP Marketing at Hubspot: a simple CMO dashboard he recommends, given the playbook shift in this AI era:

My comments:

What's not new, but is important:

• Tracking how you perform across all stages: from awareness, reach and initial engagement (i.e. at the top of funnel), to generating demand (note: pipeline is missing, but I’m sure this is a simplified view), and at capturing demand and enabling sales (revenue)

• Deeply caring about that last stage. Good marketing teams track revenue, not just awareness and MQLs/leads. That includes influencing win rates, accelerating deal velocity, and improving sales productivity.

What's new + what I like (all on the left column):

• Tracking your share of voice in AI answers (where discovery to preference happens more and more).

• Tracking creator or influencer marketing (in some way). That's a playbook that every B2B marketing team needs to master today.

💡 What does it take to win in AI search?

Great AEO resource from Josh Grant, VP Growth at Webflow, who published his definitive 2026 guide to AEO.

😍 Marketing Inspiration: Granola’s year in review

Yes, Spotify greatly improved their Spotify Wrapped after a total dud last year, but the winner - so far at least - is Granola.

Their Granola Crunched is exceptionally good marketing. It's funny and witty, and it completely strokes your ego.

They turned the temperature way up (in LLM-speak, this means increasing the model's creativity, diversity, and randomness, making outputs more varied and surprising).

It's so flattering, it makes you feel like a hero.

No surprise folks are noticing and sharing.

And I guess it worked on me too… 👇

PS: Granola is an AI note-taking app. I use it to transcribe all my meetings. That's why it's got so much context and it's becoming so valuable

👀 Stop closing new feature education overlays & tooltips in Google products (actually, most products now…)

I want to kick myself every time I close one before reading or watching.

With AI, every product improvement these days is not only super easy to use but can also be a massive time-saver.

Google Product and Eng teams are - finally! - integrating their great models deeply in all the tools where we spend much time: Search, Gmail, Slides, Drive, etc

Here’s an example for docs in G-drive that is so helpful:

See that catch-up capsule below? When you click on it, it generates a summary of the changes in that doc since you've been in there. So helpful when you collaborate with other folks.

By the way, if you're curious, MSN stands for Must-have, Should-have, Nice-to-have. These are requirements for a CMO I'm helping recruit

And here's another great update that brings together Google’s thinking and image generation models for delightful experiences. 👇

🍌 Google Slides and NotebookLM get big image and slide generation upgrades with Nano Banana Pro

If you use Google Slides, unless you're on a free plan*, you've probably noticed this magical button at the bottom:

The output is impressive. Without a single prompt, it immediately creates a slide, that uses the precise text on my slide (as you know, that was a challenge for most models until Nano Banana Pro), roughly in the same style as my deck.

I can't wait to be able to edit the output text or visuals, just like Gamma.app lets you.

Google also enabled Banana Pro in NotebookLM, allowing it to create decent infographics and slides from your notebook's content.

More details from the Google team:

“Google Slides:

• Infographic & Images: Generate detailed, professional infographics or images directly in Slides, with a simple prompt. Nano Banana Pro is connected to Google Search’s vast knowledge base, meaning can represent real-world objects and places, all while taking into account the details from your prompt.

• Beautify this slide: In just a click, turns your ideas into a visual with more accurate, legible text while modeling the look and feel of the overall deck.

NotebookLM: 

With Nano Banana Pro in NotebookLM, visualize key insights from your sources as infographics or a single, high-impact visual. You can also create complete and polished slide decks directly from your sources and share them as a PDF. See it in action.”

*available in Pro, as well as Enterprise, Business, and Education accounts. Along with the extra storage you get, the $20/month Pro tier is well worth it for the extra AI capabilities and credits.

Final Words

Thanks for sharing these highlights with busy marketing execs around you.🙏 

Someone forwarded you this email? You can subscribe here.

François | LinkedIn 

I'm a CMO, advisor, and "CMO Wingman". Yes, that's a thing :-). Ask my clients: in this AI era, CMOs need a strategic proactive advisor more than ever. I’m former CMO at Twilio, Augment Code, Apollo GraphQL, Decibel, Udacity and Head of Marketing for LinkedIn Talent Solutions.