Behind the Brief | Honest Stories from the World of Marketing

Most AI Messaging Feels Vague Because We’re Still Figuring Out the Shift

9 min · 29 de may de 2026
Portada del episodio Most AI Messaging Feels Vague Because We’re Still Figuring Out the Shift

Descripción

And the thing is, I don’t even think the products are bad. I think product marketing just hasn’t fully caught up with what AI products are becoming. Over the last few months, I’ve noticed the same terms showing up everywhere:AI agents.MCP.Agentic workflows.AI-native systems. Everyone is using the language.Very few people are explaining what actually changes because of it. So I recorded a Behind the Brief episode as a bit of an experiment, partly to simplify these ideas for other PMMs, but honestly also to process what I’ve been noticing myself. Because I think we’re entering a pretty major positioning shift:AI-assisted → AI-native. And that changes the messaging game entirely. For a long time, AI products were positioned like helpers:“AI makes your workflow faster.”“AI helps your team move quicker.”“AI assists with execution.” But AI-native systems are different. The AI isn’t helping the workflow anymore.The AI is becoming the workflow. And buyers can feel that shift, even if they don’t always have the language for it yet. That’s why vague messaging is starting to break down. Buyers are asking sharper questions now: * What does the AI actually do? * What data does it touch? * What happens when it fails? * How autonomous is it really? And honestly, I think a lot of AI jargon right now is just positioning confusion wearing technical clothing. One thing that’s become very obvious to me from working in cybersecurity and AI product marketing is this: Invisible systems only work when buyers trust them. Nobody can “see” how your AI model works.They can’t see the decision logic.They can’t see the training data. So messaging becomes the trust layer. And that’s why “AI-powered” by itself is no longer enough. One simple exercise I’d recommend: Open your homepage and look at every place you use the word “AI.” Then ask: Does the next sentence clearly explain: * what the AI is doing, * for whom, * and what changes because of it? If not, your buyers are probably already feeling that gap. I unpacked all of this more deeply in the latest Behind the Brief episode, including: * AI agents * MCP (Model Context Protocol) * Agentic AI vs generative AI * AI washing * And why positioning is becoming the competitive advantage in AI Curious what’s showing up in your world right now. What’s one AI term everyone keeps using in meetings that nobody actually explains properly? Reply and let me know, I might decode the next one. Watch the full video explanation here: Get full access to The PMM Brief at pmmshruti.substack.com/subscribe [https://pmmshruti.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

Comentarios

0

Sé la primera persona en comentar

¡Regístrate ahora y únete a la comunidad de Behind the Brief | Honest Stories from the World of Marketing!

Prueba gratis

Empieza 7 días de prueba

$99 / mes después de la prueba. · Cancela cuando quieras

  • Podcasts solo en Podimo
  • 20 horas de audiolibros al mes
  • Podcast gratuitos

Todos los episodios

31 episodios

Portada del episodio What a founder taught me about the problem PMMs keep avoiding

What a founder taught me about the problem PMMs keep avoiding

I’ve done win-loss work. So I went into this conversation half-expecting to nod along politely and half-expecting to disagree. Instead I came out with a cleaner way of thinking about a problem I’ve been circling for years. Shayaan Jagtap [https://www.linkedin.com/in/shayaanjagtap/] started Opptrack [https://www.opptrack.com/] about eight months ago, with an old coworker and a hunch about competitive intelligence. It began as a LinkedIn scraper. It’s now a win-loss engine built off more than 120 product marketing interviews. What I loved about this one is that he didn’t sell me the polished version. He walked me through the tumbling, evolving, occasionally embarrassing middle. Which is exactly what Founder Stories is for. Five things I took away: 1. Win-loss is a prioritization problem before it’s a data problem.You lost 100 deals. You can’t interview 100 people. The one-call no isn’t the same as the six-figure loss at the finish line. Knowing which losses to dig into is half the battle, and it’s the half nobody talks about. 2. More than half your losses are already in your calls.This reframed things for me. You don’t have to start with surveys and gift-card bribery. A big chunk of your losses is explainable from sales calls and CRM data you already own. Interviews are for the decisions that happened off-call, behind closed doors. 3. A loss is easier to explain than a win.His line: a win can be complex, but a loss usually comes down to one or two critical things. Pricing. Timing. A missing SOC 2. Find those and you’ve explained the deal. (His dating analogy for this was rough, and he knew it. It still landed.) 4. Saying no is the founder skill.Ten good ideas, and you have to kill nine. Every user conversation spawns more features you could build. The discipline isn’t having ideas. It’s throwing away the month of work that isn’t earning its place, and not getting emotionally married to it. 5. The AI-slop panic is punishing the wrong people.He got banned from four subreddits for posts he wrote by hand. The genuine slop gets through. We both felt this one. The old distribution playbooks are creaking, and the detection is catching real voices while missing the machines. What survives, we agreed, is a genuine voice. You can feel when a person actually wrote the thing. Why you should care:If you’re a PMM doing win-loss manually, this is a map of where the real leverage is: stop trying to interview everyone, mine what you already have, and reserve human conversations for the deals that went dark. If you’re a founder, it’s a clean look at what the unglamorous middle actually feels like. Prefer watching? Watch on YouTube. Here: Prefer listening? Listen on Spotify or Apple Podcasts. Until next time: keep learning, keep experimenting, stay curious.Shruti Get full access to The PMM Brief at pmmshruti.substack.com/subscribe [https://pmmshruti.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

17 de jul de 202639 min
Portada del episodio "Who is still sitting and writing that positioning doc? It is still me." — Swati

"Who is still sitting and writing that positioning doc? It is still me." — Swati

The PMM Skill AI Still Can’t Automate A few weeks ago I sat down with Swati, an AI-systems-focused product marketer who’s scaled B2B SaaS portfolios from $50M to $90M in ARR. I went into the conversation half-convinced I already knew what she’d say about AI. I didn’t. The gathering work is gone. The judgment work isn’t. She described what her job used to look like before agents: one human being, going through Power BI and Salesforce and email, downloading a thousand spreadsheets and trying to force them into something that made sense. “There’s only one Swati and there’s only one brain,” she said. “Every Excel requires one brain.” Now that same gathering work, pulling from Jira, from old positioning docs, from prior sales calls, gets assembled into a first draft of a battlecard in the time it used to take her just to find the right file. That part of the job, she said plainly, is automated now. But she drew a hard line right after. Positioning itself, the actual decision about what a product means to a specific buyer, is still human centric. An agent can summarize a customer call. It can’t sit across from that customer and notice the thing they didn’t quite say out loud. That’s still on her. Sales enablement is where PMM’s fingerprints stay visible The part I keep coming back to is her answer to a question most PMMs quietly struggle with: how do you prove any of this is worth the money? Her answer wasn’t the website. It was sales. In enterprise B2B SaaS, almost nobody buys off the homepage. There’s a full sales cycle, and everything inside it, down to the one-line cadence email a rep sends to a prospect, runs through product marketing. “They can’t even send an email if I don’t write it correctly,” she said, half-joking, mostly not. Content marketing and demand gen teams often end up rewriting that same positioning into blog posts that collect their own impressions and their own credit, which makes it genuinely hard to trace PMM’s fingerprints on revenue through a website. Sales enablement doesn’t have that problem. If the sales team is closing faster and closing with confidence, that’s a straight line back to the work. What this means if you’re building a personal brand while doing all of this The second half of our conversation had less to do with product marketing and more to do with the mindset it takes to post consistently while doing this job. Swati’s framing was simple: building in public is a mindset problem before it’s a tools problem. Everyone goes through a period where the content is genuinely bad, and the people who keep going are the ones who’ve made peace with that period, not the ones who found a better app. One reframe she uses when she catches herself over-scrutinizing her own content: imagine a loved one posted the exact same thing. Would you be that harsh with them? Almost never. It’s a small shift. But it’s the kind of small shift that actually changes behavior, which is rarer than it should be in advice like this. My Take Here’s what this means for product marketers: if you’re feeling replaceable right now because an agent can draft a battlecard in ten seconds, you’re measuring yourself against the wrong part of the job. The drafting was never the moat. The judgment about what goes into that battlecard, and the relationships that make sales actually trust it, are. Here’s how founders can actually apply this: stop asking your PMM to prove ROI through the website. Ask your sales team how fast they’re closing and how confident they sound doing it. That number was always the real scoreboard. We just kept looking at the wrong dashboard. Curiosity Corner If we had another hour... * I’d ask her to walk through one specific battlecard, start to finish, and show exactly where the agent’s draft ended and her judgment took over * I’d explore how she measures “sales confidence” in something other than a gut feeling, is there an actual signal she tracks * I’d challenge her on whether the loved-one reframe holds up under real public criticism, not just her own self-doubt Full conversation with Swati is up now on YouTube, Spotify, and Apple Podcasts. If you got something out of this, forward it to a PMM who needs to hear it. Get full access to The PMM Brief at pmmshruti.substack.com/subscribe [https://pmmshruti.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

3 de jul de 20261 h 0 min
Portada del episodio 100,000 USERS LATER...

100,000 USERS LATER...

Most people think great SaaS products start with a brilliant idea. Joseph Lee’s started with a customer saying: “Your website wasn’t very clear... but now I get it.” That moment changed everything. On the latest episode of Behind the Brief, I sat down with Joseph, CEO & Co-Founder of Supademo, to unpack how one customer interaction turned into one of G2’s fastest-growing products. A few takeaways that stayed with me: → Customers don’t buy features. They buy understanding. → Product demos shouldn’t just tell. They should show. → Growth rarely comes from one channel. It’s the result of hundreds of small decisions executed consistently. → Storytelling is becoming one of the biggest competitive advantages in SaaS. One insight I especially loved: Instead of pitching prospects, Joseph’s team would build a demo for them first. No hard sell. No aggressive outreach. Just value. That simple shift helped create trust before asking for anything in return. And maybe the most important reminder for anyone building in public: Perfection is overrated.Consistency wins. Listen to the full conversation below 👇 What’s one marketing lesson you’ve learned that only became obvious after doing the work? #BehindTheBrief #FounderSeries #ProductMarketing #SaaS #Startups #BuildInPublic #MarketingLeadership Get full access to The PMM Brief at pmmshruti.substack.com/subscribe [https://pmmshruti.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

10 de jun de 202619 min
Portada del episodio Most AI Messaging Feels Vague Because We’re Still Figuring Out the Shift

Most AI Messaging Feels Vague Because We’re Still Figuring Out the Shift

And the thing is, I don’t even think the products are bad. I think product marketing just hasn’t fully caught up with what AI products are becoming. Over the last few months, I’ve noticed the same terms showing up everywhere:AI agents.MCP.Agentic workflows.AI-native systems. Everyone is using the language.Very few people are explaining what actually changes because of it. So I recorded a Behind the Brief episode as a bit of an experiment, partly to simplify these ideas for other PMMs, but honestly also to process what I’ve been noticing myself. Because I think we’re entering a pretty major positioning shift:AI-assisted → AI-native. And that changes the messaging game entirely. For a long time, AI products were positioned like helpers:“AI makes your workflow faster.”“AI helps your team move quicker.”“AI assists with execution.” But AI-native systems are different. The AI isn’t helping the workflow anymore.The AI is becoming the workflow. And buyers can feel that shift, even if they don’t always have the language for it yet. That’s why vague messaging is starting to break down. Buyers are asking sharper questions now: * What does the AI actually do? * What data does it touch? * What happens when it fails? * How autonomous is it really? And honestly, I think a lot of AI jargon right now is just positioning confusion wearing technical clothing. One thing that’s become very obvious to me from working in cybersecurity and AI product marketing is this: Invisible systems only work when buyers trust them. Nobody can “see” how your AI model works.They can’t see the decision logic.They can’t see the training data. So messaging becomes the trust layer. And that’s why “AI-powered” by itself is no longer enough. One simple exercise I’d recommend: Open your homepage and look at every place you use the word “AI.” Then ask: Does the next sentence clearly explain: * what the AI is doing, * for whom, * and what changes because of it? If not, your buyers are probably already feeling that gap. I unpacked all of this more deeply in the latest Behind the Brief episode, including: * AI agents * MCP (Model Context Protocol) * Agentic AI vs generative AI * AI washing * And why positioning is becoming the competitive advantage in AI Curious what’s showing up in your world right now. What’s one AI term everyone keeps using in meetings that nobody actually explains properly? Reply and let me know, I might decode the next one. Watch the full video explanation here: Get full access to The PMM Brief at pmmshruti.substack.com/subscribe [https://pmmshruti.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

29 de may de 20269 min
Portada del episodio What 700 → 90K Followers Taught Me About Consistency (and Money)

What 700 → 90K Followers Taught Me About Consistency (and Money)

We spent an hour talking about Instagram. But not the kind you usually hear about. No “post 5 times a day.”No “use this hashtag.”No “hack the algorithm.” Instead, we spoke about: * loneliness while building something new * the fear of putting yourself out there * and why most creators never make money One thing stood out to me: Consistency is not glamorous.It’s repetitive. It’s boring.And most of the time, it feels like nothing is working. Until suddenly; it does. If you’re someone sitting on drafts right now…this might be your sign to just hit publish. 🎧 Full episode: Get full access to The PMM Brief at pmmshruti.substack.com/subscribe [https://pmmshruti.substack.com/subscribe?utm_medium=podcast&utm_campaign=CTA_4]

11 de abr de 202644 min