Best of LinkedIn: AI in B2B Marketing CW 51 - 02

00:00:00: This episode is provided by Thomas Allgeier and Frennis, based on the most relevant LinkedIn posts about AI and BDB marketing in calendar weeks fifty-one to two.

00:00:09: Frennis is a B to B market research company helping enterprise marketing teams sharpen their strategies and outreach with customer segmentation, ideal customer profiles and deep dives, customer needs analyses, and buying center insights.

00:00:23: And looking at those sources, you know, all the LinkedIn conversations from the end of last year into the new one, something really stands out.

00:00:30: That's it.

00:00:31: We've finally, and I mean finally, moved past the generic AI hype cycle.

00:00:36: The whole conversation has shifted.

00:00:38: It's not

00:00:38: about what if.

00:00:39: Not at all.

00:00:40: It's a completely focused on actual concrete go-to-market execution on strategy.

00:00:45: Exactly.

00:00:45: It's all about the operational reality now.

00:00:47: So for this deep dive, our mission is to pull out the top AI trends that are really impacting strategic B to B marketers, like right now.

00:00:54: And we need to get past the buzzwords, right?

00:00:56: Oh, completely.

00:00:56: We've got to figure out what's actually working, what's generating revenue, where the big failures are, and maybe the biggest question of all, how the human role is changing.

00:01:06: Yeah, how we fit in alongside all these autonomous systems.

00:01:10: We've got five major themes to get through.

00:01:12: So let's start at the very top.

00:01:13: Let's talk.

00:01:14: AI strategy and operating models.

00:01:16: Makes sense.

00:01:17: You have to start with strategy because, I mean, that's where the money's going.

00:01:21: And Fleet Britons pointed out a really key shift in how executives are thinking.

00:01:25: They're not just looking at it as a cost cutting tool anymore.

00:01:28: No, not at all.

00:01:29: They see AI as a growth lever.

00:01:32: The priority now is massive revenue and valuation uplift.

00:01:38: not just trimming the budget.

00:01:39: That's a fascinating change.

00:01:40: It feels like they've moved from just optimizing a spreadsheet to fundamentally changing how they win market share.

00:01:47: It's a much bigger bet.

00:01:48: Way

00:01:49: bigger, riskier for sure, but the potential reward is so much higher than just shaving a few dollars off your sauce bill.

00:01:55: It is, but with that kind of ambition, you need discipline.

00:01:59: You need diagnosis.

00:02:00: That's what David Brown was really hammering home.

00:02:03: Diagnose first?

00:02:04: Yes.

00:02:04: Diagnose your GTM friction, your actual business needs before you start funding a bunch of scattered AI experiments and just hoping one works.

00:02:11: So what happens if you skip that step?

00:02:13: If your team just gets excited and buys a shiny new tool,

00:02:17: you just accelerate chaos.

00:02:18: Right.

00:02:19: Robert Buchitz had this great analogy that really stuck with me.

00:02:22: He said, AI basically accelerates whatever GTM system you already have in place.

00:02:27: He compared it to a Rubik's Cube.

00:02:28: I think I saw that one.

00:02:30: If the foundation is already a mess, AI just makes it a bigger mess

00:02:34: faster.

00:02:34: Precisely.

00:02:35: If your GTM is scrambled bad targeting, broken sales handoffs, whatever AI doesn't magically fix it.

00:02:43: It just multiplies the error rate.

00:02:44: It executes your bad strategy with machine speed.

00:02:47: And that's not transformation, that's just catastrophe on an accelerated timeline.

00:02:52: And we're seeing the data to back up that idea of catastrophe, which honestly should be a huge warning for anyone rushing into pilots right now.

00:03:00: The numbers are pretty stark.

00:03:01: I mean Robert Buchitz cited research showing that roughly ninety-five percent of generative AI pilots show no measurable PNL impact.

00:03:07: Ninety-five

00:03:08: percent, that is a shocking number.

00:03:09: And it makes you wonder, right, the tech just fundamentally flawed or are we just using it all wrong?

00:03:14: That's

00:03:15: the question.

00:03:15: I mean, are we sure that failure rate isn't just because they picked the wrong KPIs or they were looking for ROI in the wrong places?

00:03:22: Because, ninety-five percent failure, that's just widespread waste.

00:03:25: It's wasted effort for sure, but it's not wasted technology.

00:03:30: The sources are really consistent on this.

00:03:32: In like, seventy to eighty percent of those failures, it all comes down to execution friction.

00:03:37: So, poor data, no clear owner.

00:03:40: Misaligned processes.

00:03:41: The tech itself isn't weak.

00:03:42: It's the foundation that's being built on.

00:03:44: So it's not an LLM problem.

00:03:46: It's a CRM problem.

00:03:47: Exactly.

00:03:48: And to really drive that point home, Greg Hansen brought up some MIT data.

00:03:52: It contrasted generic AI projects with specialized ones.

00:03:55: And the difference was?

00:03:56: Night

00:03:56: and day.

00:03:57: The generic broad use projects, they succeed only five percent of the time.

00:04:01: But the specialized ones, the ones built to solve one very specific GTM problem, they succeed sixty seven percent of the time.

00:04:07: Wow.

00:04:08: That completely changes the math for B to B leaders.

00:04:11: It's not about finding one tool to do everything.

00:04:13: No, it's about investing in hyper niche problem specific AI.

00:04:17: So the big takeaway here then is that operationalizing AI now is critical.

00:04:22: I think Mahesh Iyer said that.

00:04:23: He did.

00:04:24: And David Edelman strongly reinforced it.

00:04:26: The real competitive advantage won't come from just having the newest model.

00:04:31: It's going to be one by your operating model.

00:04:33: The systems, the data, the process.

00:04:36: The system beats the tool.

00:04:37: Okay, that makes perfect sense.

00:04:39: Let's move from that high-level strategy down to the execution layer.

00:04:43: that's seen, well, frankly, the most hype and the most failure.

00:04:48: AISDRs.

00:04:49: AISDRs

00:04:50: and sales automation.

00:04:51: Okay, before we get into it, we have to address the elephant in the room.

00:04:55: The AISDR was supposed to be the replacement.

00:04:57: for human outreach.

00:04:58: Is that actually what's happening?

00:05:00: It's

00:05:00: a lot more nuanced than that.

00:05:01: The way they're being positioned now is more like specialized teammates for qualification and outbound, not wholesale replacements.

00:05:08: So they're part of the team.

00:05:09: Exactly.

00:05:09: Yeah.

00:05:09: I mean, Liam Sheridan mentioned tools like Gemini III for cold outreach.

00:05:13: And Roman Cerny was talking about Gojbury AI for booking demos.

00:05:17: The advocates, they really believe these agents will be a mandatory part of the tech stack very, very soon.

00:05:22: And that potential is there.

00:05:24: But let's talk about the immediate widespread mistake everyone made.

00:05:27: Jason M. Lemkin called it out perfectly, which was, so many companies just bought an AI SDR tool and turned it on just like that.

00:05:35: No thought about training or segmentation or integration.

00:05:39: And

00:05:39: the results of that spray and pray automation strategy were

00:05:43: exactly what you'd expect.

00:05:44: Predictable failure and a whole lot of spam.

00:05:48: Right.

00:05:48: Kyle Assay saw failures because of just irrelevant personalization.

00:05:53: You know, the AI emailing.

00:05:55: existing customers instead of prospects, the response rates were near zero.

00:05:59: I can imagine.

00:06:00: And Steven M. Lois and Matt Lush were even more blunt about it.

00:06:03: They call them glorified mail merge, just automating spam at scale.

00:06:06: Scaling

00:06:07: mediocrity just gives you more mediocre results, but faster.

00:06:10: So AI isn't the problem, but the old volume based strategy is just dead.

00:06:14: What's the successful counter strategy then?

00:06:16: Precision.

00:06:17: It's all about quality over volume now.

00:06:19: Using AI as a filter, not just a broadcaster.

00:06:22: Okay,

00:06:22: tell me more.

00:06:23: Kyle Isay shared this great success story.

00:06:25: that really shows the flip.

00:06:27: They started with a list of one thousand accounts.

00:06:29: But instead of just blasting all one thousand... No

00:06:31: didn't.

00:06:32: No.

00:06:33: They used the AI to narrow that list down to just fifty.

00:06:37: The absolute best fit, most qualified accounts for their human AEs to focus on.

00:06:42: So the

00:06:42: AI isn't a scaling tool for quantity, it's a precision filter for quality.

00:06:47: That is a fundamental change in how GTM has worked for what, a decade?

00:06:51: It is.

00:06:51: and the outcome.

00:06:52: Let

00:06:52: me guess, they got meetings.

00:06:54: Immediate qualified meetings.

00:06:55: They used AI to go narrow.

00:06:57: not wide.

00:06:58: And that kind of specificity must require some pretty complex orchestration behind the scenes.

00:07:03: It does.

00:07:04: Alan Ruckstein laid out a blueprint for it.

00:07:06: He suggests mapping your entire SDR workflow and then assigning each specific task to its own AI agent.

00:07:12: It's not one AI SDR, it's a whole modular team.

00:07:16: Like a micro team of specialists.

00:07:18: Exactly.

00:07:19: You have a research agent digging up Intel, a CRM agent logging everything and keeping data clean, and then a sequencer agent.

00:07:25: that actually manages the outreach.

00:07:26: So

00:07:27: you're replacing what slows reps down, the boring repetitive stuff, not the reps themselves.

00:07:31: That's the goal.

00:07:33: Free them up for empathy for judgment.

00:07:35: And Ted Carter even showed off a GTM team he built with sixteen different AI agents, all focused purely on precise account qualification.

00:07:43: Sixteen.

00:07:44: That's like a digital orchestra.

00:07:46: It is.

00:07:47: And of course an orchestra like that, with sixteen different players, it completely falls apart if you don't have the right fuel.

00:07:54: Which brings us to our third theme

00:07:55: the fuel.

00:07:56: Yeah

00:07:57: data infrastructure and measurement This

00:07:59: has to be the hidden ROI killer.

00:08:01: I mean Amon Carrour Sandeep Gulati.

00:08:03: They all confirmed it.

00:08:05: poor data quality and fragmented systems are the number one reason these AI projects fail,

00:08:10: right?

00:08:10: If the data is messy, the AI is basically just doomed to automate mediocrity.

00:08:15: That's exactly what Sandeep Gulotti warned.

00:08:17: He said, fix your data hygiene before you invest in models, otherwise you're just pouring rocket fuel on a pile of garbage.

00:08:22: It's a perfect analogy.

00:08:23: So for people listening who are trying to integrate these new tools with their legacy systems, what's the architectural advice that's coming

00:08:30: out?

00:08:31: Jesper Ruiz had some specific advice for anyone building a complex GTM system.

00:08:36: He recommended using a technical backbone like SuperBase for things like lead scoring, something with real SQL capabilities.

00:08:43: Over simpler tools like air table.

00:08:45: Right.

00:08:45: You just need that database integrity and speed to feed these complex multi-agent systems we're talking about.

00:08:51: And that points to a bigger shift in Martek, doesn't it?

00:08:53: I'm seeing a lot of chatter that the role of the CRM itself is changing.

00:08:58: It absolutely is.

00:08:59: Steven Beis and Gal Aga both made the argument that the CRM is becoming more of a passive data store.

00:09:04: It's just the plumbing.

00:09:05: It's not where the work happens anymore.

00:09:07: No.

00:09:08: The real-time deal work, the actual execution that's moving to newer platforms where these AI agents can be autonomous and act on data instantly.

00:09:17: The back office isn't where the action is.

00:09:19: So if AI is changing how we work and where we work, it has to change how we measure success, right?

00:09:24: What are the new metrics leaders are focusing on?

00:09:26: Matthew Thompson noted a big shift toward two key ratios, efficiency and productivity.

00:09:32: Okay,

00:09:32: break those down.

00:09:33: Efficiency is net new ARR generated per dollar spent.

00:09:37: Simple as that.

00:09:38: Are you getting a good return on the cost of the agents and the compute time?

00:09:42: and productivity.

00:09:43: Productivity

00:09:44: is net new ARR per GTM full-time employee because if you've invested all this money in AI and your human team isn't generating significantly more revenue per person.

00:09:54: Then what's the point?

00:09:55: Exactly.

00:09:56: The investment is just adding cost without giving you any strategic advantage.

00:10:00: Those are the two numbers AI has to move.

00:10:02: Clear, quantifiable, love it.

00:10:04: Okay, let's pivot from the internal stuff to external visibility and content.

00:10:08: Theme four, content, visibility, and the human rule.

00:10:12: This feels like where things are really transforming.

00:10:14: This

00:10:14: is probably the biggest philosophical change for content marketers.

00:10:18: Ryan Yackey called AI the new discovery engine.

00:10:20: We are moving beyond just optimizing for a search bar ranking.

00:10:24: So, beyond traditional SEO.

00:10:26: Way beyond.

00:10:27: Marcus Sheridan actually predicted that AEO agentic engine optimization is going to be seen as more important than SEO and soon.

00:10:34: Why is that?

00:10:35: Because the AI doesn't just care if you're ranked highly.

00:10:37: It rewards you for being cited and trusted as the authoritative source by the models themselves.

00:10:43: So, AEO is about optimizing for machine comprehension, not just for human clicks.

00:10:49: Your content has to be so good, so structured, that an LLM will actually cite you as the source of truth in its answers.

00:10:56: That is a perfect summary.

00:10:58: And because these AI systems pull signals from all over the place, Sarah Stella Latanzio said that ignoring things like brand mentions is basically self-sabotage.

00:11:06: That completely changes your content strategy.

00:11:09: It's not about one web page anymore.

00:11:10: It's about building authority across the entire digital ecosystem.

00:11:14: Which is why Chase Diamond pointed out that AI pulls signals from places we used to overlook.

00:11:19: Reddit threads, forums, customer reviews.

00:11:21: It's looking for real unfiltered signals, not just the polished corporate blog posts.

00:11:26: And this even applies to video, which is a huge deal for B to B education.

00:11:30: Thomas Ross had a very clear warning about this.

00:11:33: He did.

00:11:33: He said, remember, AI doesn't watch your videos.

00:11:36: It reads the transcript.

00:11:37: The optimized transcript is the core source of visibility for AEO.

00:11:40: If you're not optimizing that, your video content is essentially invisible to these agents.

00:11:45: So as the machines get better at all this, what happens to the human marketer?

00:11:50: Is our judgment still the most valuable thing we have?

00:11:52: Absolutely.

00:11:53: Alex Baca pointed this out.

00:11:54: AI is a productivity accelerator, but mostly for experts.

00:11:59: It's actually widening the gap with junior workers who don't have the critical thinking to use the tools well.

00:12:04: So critical thinking is the new competitive edge.

00:12:07: You heard your schwaner said that's exactly it.

00:12:09: You have to be the human in the loop.

00:12:11: You have to supervise, question, and stress test what the AI gives you.

00:12:15: If you can't spot a subtle mistake, the tool just makes you faster at being mediocre.

00:12:20: But how do you protect the unique parts of your brand?

00:12:24: The soul, the voice, the stuff you can't automate.

00:12:27: That is the critical human differentiator.

00:12:30: Carolyn Healy shared this powerful case study where a team automated its brand's soul and engagement just plummeted by forty percent.

00:12:37: When everything starts to sound the same, that human premium branding, that authenticity, that becomes your key differentiator.

00:12:45: The whole point is to free up our time for empathy, for relationships, not to replace deep thinking.

00:12:50: That's a

00:12:51: powerful warning.

00:12:53: Okay, let's move to our final theme.

00:12:55: Ecosystem risk and guardrails.

00:12:57: The market is definitely maturing.

00:12:58: Yeah, we're seeing major validation.

00:13:00: Alina Vandenberg pointed to Salesforce's acquisition of Qualified as a huge signal.

00:13:05: It's boosting awareness around inbound AISDRs in a big way.

00:13:09: The toolkit is growing up.

00:13:11: And

00:13:11: with powerful new tools come powerful new risks.

00:13:14: especially around data privacy.

00:13:15: We

00:13:15: need guardrails.

00:13:17: Nicole Leffer gave a specific warning about consumer tools, reminding everyone to make sure that improve the model for everyone.

00:13:23: Toggle is switched off in ChatGPT if you're handling any proprietary data.

00:13:26: And what about internal risk management?

00:13:29: Arkanevolvianathan said you need two distinct AI zones.

00:13:32: First, an innovation zone where risk is low and you can experiment.

00:13:36: But second, you need a revenue execution zone where the guardrail's prompts, model settings, human review have to be incredibly strict.

00:13:44: non-negotiable.

00:13:44: That's

00:13:45: a really practical framework.

00:13:46: Let's finish up with the financial side of things.

00:13:48: Jason M. Lemkin noted that AI is really changing how we perceive pricing.

00:13:53: It is.

00:13:54: It's a paradox.

00:13:55: On one hand, replacing ten SDRs with a fifty thousand dollar AI agent seems cheap.

00:14:01: Right.

00:14:01: But on the other hand, we're starting to expect core AI features to just be included in our existing platforms.

00:14:08: So a small extra subscription fee for some minor AI feature can suddenly feel really expensive.

00:14:15: It's the paradox of commoditization.

00:14:16: We just expect the intelligence to be baked in now.

00:14:19: Exactly.

00:14:20: And Terry Zellin reminds us that the winning AI marketing stacks aren't the ones with the most tools.

00:14:25: They're the ones that clarify the connections between systems and centralize the logic.

00:14:30: which brings us right back to where we started.

00:14:32: System architecture and strategy.

00:14:34: If you enjoyed this deep dive, new episodes drop every two weeks.

00:14:37: Also, check out our other editions on account-based marketing, field marketing, channel marketing, martech, go-to-market, and social selling.

00:14:44: Thanks for tuning in and sharing this with us.

00:14:46: And remember to subscribe so you don't miss the next deep dive into the B to B world.

00:14:50: And as you think about this huge shift in GTM strategy, here's a final thought for you based on an insight from Shashi Bellamkanda.

00:14:57: This move toward agentic engine optimization and AI-driven commerce means your traditional corporate website is becoming secondary.

00:15:04: It's being replaced by user intent as AI agents handle the discovery and transaction logic.

00:15:09: Shashi Bellamkanda called this the faceless economy.

00:15:13: So what does it mean for your brand when your primary customer is no longer a human browsing your site, but an algorithm citing your content?