Every week, we track dozens of stories at the intersection of AI and product marketing. Most are noise. A few are signals. And occasionally, one number lands that should change how you think about your job.
This week had that number.
That's not a trend line. That's a structural shift. Nearly half of your buyers are forming opinions about your product in ChatGPT, Claude, or Perplexity before they ever visit your website.
Combined with a parallel stat — 64% of consumers now use AI for purchase research — we're watching the buyer's journey fundamentally rewire in real time.
Here are the five stories that matter most, stack-ranked by their impact on product marketing.
The Stack Rank
March 11, 2026
42% of B2B Buyers Now Start Research in AI Search
The killer stat. This isn't about adding a new channel — it's about the first touch point shifting away from your controlled properties. If nearly half your buyers form initial opinions before they see your messaging, your content strategy must change.
The implications are immediate: How does your product appear when someone asks Claude or Perplexity about solutions in your category? Do you even know? Most PMMs don't.
This is the core argument for writing for two audiences — humans who read and AI systems that parse.
→ Direct PMM action requiredMarch 12, 2026
Salesforce Renames Marketing Cloud to "Agentforce Marketing"
This isn't a rebrand — it's a strategic re-architecture. Salesforce is rebuilding Marketing Cloud from the ground up for agentic AI.
The shift: from one-way campaigns to two-way conversations. The "do not reply" era is officially dead. When a prospect replies to a marketing email, an AI agent can now recognize intent, answer questions, and route to sales appropriately.
Marketing becomes dialogue, not broadcast. When the largest marketing cloud vendor makes this move, it signals where the entire category is heading.
→ Category-defining moveMarch 12, 2026
MarTech: Marketing Needs "Decision Infrastructure" for AI
The sleeper hit of the week. This piece explains why AI crushes it in code but struggles in marketing.
The answer: code has syntax, structure, and version control. Marketing runs on Slack threads, gut instinct, and "that leader with a strong opinion."
McKinsey estimates marketing and sales will capture $400–660 billion annually from generative AI — but only if teams build the missing infrastructure. Your marketing logic needs to be codified into structures AI can actually use. Context graphs and documented decision frameworks become competitive advantages.
Here's the uncomfortable truth: most martech stacks can't deliver this. Decision infrastructure requires connected data across ERP, CDP, and CX — not siloed point solutions. The vendors with true end-to-end architectures (think embedded customer data within ERP, connected to commerce, service, and marketing) have a structural advantage. Everyone else is duct-taping APIs together. (I wrote about why agents will finally force ERP and CX to talk — this is the same argument.)
This is the "why" behind most AI readiness gaps.
→ The "why" behind AI readiness gapsMarch 7, 2026
64% of Consumers Use AI Tools for Purchase Research
The B2C companion to our #1 story. Combined, these stats show the buyer's journey has fundamentally changed across both B2B and B2C.
Traditional marketing funnels assumed you controlled the awareness and consideration stages through advertising and content. That model is breaking. Discovery is increasingly happening in AI interfaces where you have no direct presence.
The question isn't whether to adapt — it's how fast.
→ Consumer behavior validationMarch 11, 2026
Perplexity's "Computer" Claims to Replace $225K in Marketing Tools
Built "in a weekend," this AI assistant autonomously monitors dashboards, optimizes campaigns, and coordinates marketing activities.
The shift from AI-as-advisor to AI-as-operator is happening faster than expected. The $225K figure (likely exaggerated for effect) still makes the point: AI is moving from answering questions to executing workflows.
For PMMs, the question becomes: which parts of your job require human judgment, and which are ripe for automation? That distinction is your career moat.
→ Tool disruption signalWhat This Means for PMMs
If I had to distill this week into one insight, it's this:
💡 The Bottom Line
The buyer's journey is rewiring around AI. Nearly half of B2B buyers and two-thirds of consumers now use AI tools for research before engaging with vendors directly. Meanwhile, the tools PMMs use are being rebuilt for agentic AI (Salesforce), and the skills we've relied on are being automated faster than expected (Perplexity). The PMMs who thrive will be those who understand both shifts — adapting their content for AI discovery while identifying which human skills remain irreplaceable.
Actions for This Week
- Query AI about your product. Ask ChatGPT, Claude, and Perplexity what they know about your product and category. Document the gaps. That's your GEO roadmap.
- Audit your "decision infrastructure." How much of your marketing logic is documented vs. living in Slack and meetings? Start codifying.
- Watch the Salesforce shift. If you're on Marketing Cloud (or competing against it), understand what "Agentforce Marketing" means for your roadmap.
- Read the full analysis on writing for two audiences — it's the practical playbook for responding to the 42% stat.
📬 Get the Daily Digest
Every morning, the 5-6 stories shaping AI + PMM. Free, 2 minutes with your coffee.
Browse the Archive →