Best of LinkedIn: MarTech Insights CW 42/ 43

00:00:00: This deep dive is provided by Thomas Allgaier and Frennus, based on the most relevant LinkedIn posts about MarTech in calendar weeks forty-two and forty-three.

00:00:08: Frennus is a B to B market research company helping enterprises gain the market, customer, and competitive insights needed to drive growth and success.

00:00:16: Welcome to the deep dive.

00:00:18: So today we're diving into MarTech, looking at what's really buzzing on LinkedIn these last couple of weeks.

00:00:24: And, you know, it feels like there's this... This real shift happening.

00:00:28: Less hype, more pragmatism.

00:00:29: Definitely.

00:00:30: It's like a collective strategic reset.

00:00:32: People are moving away from just chasing the next shiny object and focusing much more on, you know, execution.

00:00:38: Making things actually work.

00:00:39: Right.

00:00:39: Fixing what's broken.

00:00:40: Exactly.

00:00:41: We've looked through a lot of posts and the big themes are pretty clear.

00:00:44: stack hygiene, this concept of egetic AI taking charge, and a serious focus on clean data.

00:00:50: The message seems to be stop piling on complexity, start orchestrating what you have and getting value.

00:00:54: Okay, let's unpack this then.

00:00:55: Where should we start?

00:00:56: Maybe with that first point, the stack hygiene.

00:00:58: Yeah, let's do that.

00:00:59: It's probably the most dominant theme.

00:01:01: this idea that you just have to clean up the existing mess.

00:01:04: Leaders are asking, finally, why they're paying for so many tools they barely use.

00:01:08: And that

00:01:08: cost, it's not just the subscription fee, is it?

00:01:11: I saw Emery Gooney asking that exact question, why keep adding tools if you maybe use, what, eight out of forty seven?

00:01:17: He made a great point that the subscription is often the cheapest part.

00:01:21: Oh, absolutely.

00:01:22: The real cost comes with, you know, the training, the maintenance, trying to make all these different tools talk to each other.

00:01:27: Right,

00:01:28: the integration nightmare.

00:01:29: And that sprawl.

00:01:30: It hits the bottom line hard.

00:01:32: Arcadia's Saradin shared some data, actually.

00:01:35: Teams overspending on tools they don't use.

00:01:37: They could be looking at a, uh, forty-eight percent higher IT costs.

00:01:41: Yeah.

00:01:41: Almost fifty percent

00:01:42: higher.

00:01:43: Wow.

00:01:43: Forty-eight percent.

00:01:44: That's huge.

00:01:45: It

00:01:45: really is.

00:01:45: It slows everything down to campaigns, results, everything.

00:01:49: So if you're thinking about upgrading a platform, like moving to a new CDP or automation tool, it's not just a lift and shift job.

00:01:57: No way.

00:01:58: Chris Stoltz warned about this.

00:02:00: If you just migrate without rethinking why you bought the original tool and what you really need for now, you're just moving, as he put it, old baggage.

00:02:09: Yeah, just putting old problems in a shiny new box

00:02:12: doesn't help.

00:02:12: Exactly.

00:02:13: You've got to fix the leaks first.

00:02:16: Sean Warren Cool was really clear on this.

00:02:19: He laid out like four foundational fixes you absolutely need before you even think about adding more complex AI.

00:02:25: Okay, what are they?

00:02:26: First, define one single source of truth for your customer data.

00:02:29: Really nail that down.

00:02:31: Second, get serious about instrumenting your first-party data, often using server-side tagging.

00:02:36: Third, actively reduce the tool sprawl, cut things out.

00:02:40: And fourth, build proper governance before you let generative AI loose.

00:02:44: That makes so much sense.

00:02:45: Get the house in order first.

00:02:47: It's not just about saving money now, it's setting yourself up for the future, right?

00:02:50: Like Paul Feldman argued, more tech modernization needs a proper multi-year strategy.

00:02:56: and alignment across the business.

00:02:57: Yeah, alignment too.

00:02:58: So the tech actually delivers a good brand experience and makes you ready for what's next, like AI.

00:03:03: Which

00:03:04: brings us neatly to that second theme, AI.

00:03:07: But not just AI as a tool, AI is something more.

00:03:11: Right, this is where things get really interesting, isn't it?

00:03:13: The conversation seems to be shifting from AI as just, you know, an assistant or an automation feature to AI becoming the main coordinator.

00:03:23: the orchestrator.

00:03:24: That's the word agentic AI.

00:03:26: We're seeing that term pop up more.

00:03:27: Well, the late Thomas Torres actually called it the agentic AI orchestration layer.

00:03:32: Think of it like the conductor for your entire MarTech stack.

00:03:35: Okay,

00:03:35: like

00:03:35: how?

00:03:36: Well, this agent layer pulls data from the CDP, grabs assets from your library, talks to APIs, uses large language models, all to deliver outcomes in real time.

00:03:46: Senoramiki agreed, saying this agentic layer is now central, whether you're using an MACA setup or a more traditional stack.

00:03:52: It's the new brain.

00:03:53: So

00:03:53: if AI is the conductor or the brain, that changes how we even think about the structure of the stack itself, doesn't it?

00:04:00: We usually picture layers, right?

00:04:01: Data at the bottom, engagement on top.

00:04:03: That static model is breaking down.

00:04:05: Florian Delval had a great take on this.

00:04:07: He said, Martek is moving away from those fixed layers towards Uh, dynamic, agentic networks.

00:04:13: Agentic networks.

00:04:14: Yeah.

00:04:14: Imagine AI agents moving across different systems, triggering actions, making decisions based on probability.

00:04:21: If the agents are everywhere doing things autonomously, then yeah, the old idea of a fixed top or bottom of the stack, it sort of dissolves.

00:04:29: That

00:04:29: sounds incredibly powerful, but also... Maybe a bit risky.

00:04:34: If these agents are running around making decisions, who's making sure they don't go off the rails?

00:04:38: Where are the guardrails?

00:04:38: Ah,

00:04:39: good question.

00:04:40: And the answer seems to be marketing operations, MOPs.

00:04:44: Jessica Kow pointed out that gen AI adoption is moving faster than governance right now.

00:04:47: Which sounds chaotic.

00:04:49: It can be.

00:04:50: So MOP's leaders are becoming crucial.

00:04:51: They're the ones who need to build the frameworks, the policies, the data discipline to scale AI safely and ethically.

00:04:57: Yeah.

00:04:57: Tarun Rathnam emphasized this too.

00:04:59: Implementation has to cover process, platforms, and people.

00:05:03: It's not just tech.

00:05:05: So the ultimate goal is still about the customer, presumably, better experiences.

00:05:09: Hopefully.

00:05:10: Andrew Nielsen made a neat distinction.

00:05:12: He said, traditional AI was mostly built to respond to things customers did.

00:05:17: Agented AI, though, is being built to foresee, to look at signals and context and anticipate what a customer might need next, deliver those anticipatory experiences.

00:05:26: OK, that's the big vision.

00:05:28: But what about the reality for companies building these AI tools, especially startups?

00:05:33: Yeah,

00:05:33: there's a harsh reality check there.

00:05:35: Aditya Venpati shared a pretty sobering statistic.

00:05:38: Something like eighty percent of the AI tools being tested get churned within a year.

00:05:42: Eighty percent?

00:05:43: Why so high?

00:05:44: Because, apparently, the senior buyers testing these tools learn enough during the pilot to figure out how to replicate the narrow function in-house.

00:05:53: The lesson for startups.

00:05:55: Don't just offer a point solution.

00:05:56: You need to deliver something deeply embedded, an end-to-end application, basically, to survive.

00:06:01: Ouch, a tough lesson in building sticky products.

00:06:04: Okay, let's shift gears a bit.

00:06:05: Theme three, data, integrity, privacy, and this convergence we're starting to see.

00:06:10: It always comes back to data, doesn't it?

00:06:11: Always.

00:06:12: And tactically, signal loss is still a major headache for marketers, you know, with browser changes and privacy measures.

00:06:18: Right.

00:06:20: highlighted Google Tag Gateway as a really practical fix for that.

00:06:24: Essentially, it lets you serve tags as first-party data, which helps preserve signals that might otherwise get blocked.

00:06:31: He mentioned seeing, on average, an eleven percent uplift in captured conversions for Google Ads users using it.

00:06:36: Okay, so that's a tangible improvement, and it underlines that point about clean data signals being more important than just raw volume, maybe.

00:06:42: Exactly.

00:06:43: Quality over quantity.

00:06:45: And AI is also helping unlock insights from data that used to be really hard to analyze at scale.

00:06:51: like qualitative data.

00:06:53: Justin Norris developed this fascinating AI workflow for win-loss analysis.

00:06:58: Instead of someone manually reading notes from a few deals, this AI goes through call transcripts, notes everything, cross maybe a thousand opportunities.

00:07:05: It then creates a standardized driver dictionary.

00:07:08: So it makes all that messy qualitative stuff quantifiable and actionable.

00:07:11: That's clever.

00:07:12: Turning anecdotes into actual data.

00:07:15: But when we talk about gathering all these signals, especially identifying users, the trust and privacy issue looms large.

00:07:21: Huge.

00:07:22: Clark Barron had a really thought-provoking take on denonymizing website visitors too aggressively.

00:07:29: He basically argued it's a form of relational coercion.

00:07:31: Relational

00:07:32: coercion?

00:07:33: How?

00:07:34: Because it claims a familiarity, it asserts access without actually earning consent or building a relationship first.

00:07:40: It makes personalization feel less like helpful service and more like... Well, surveillance.

00:07:45: J. Mandel echoed this, pushing for a clean data alliance and saying trust is really the only metric that matters long term.

00:07:51: So this push for trustworthy clean data is that helping drive some system changes too, like finally breaking down the walls between different departments

00:08:00: tech.

00:08:00: It seems like it.

00:08:01: Scott Brinker didn't mince words.

00:08:03: He called the historical split between MarTech and AdTech silly stupid and damaging to the customer experience.

00:08:09: Strong words, but probably true for many.

00:08:11: Right.

00:08:12: But he thinks AI is finally forcing these worlds together.

00:08:15: And we are seeing concrete examples.

00:08:17: Jonathan Mack highlighted the new integration between Amazon Marketing Cloud and Adobe's real-time CDP.

00:08:23: That's a big step towards unification.

00:08:26: OK, interesting.

00:08:27: So systems are converging.

00:08:28: Let's talk about the final theme.

00:08:30: architecture and results, specifically this idea of composable stacks and actually measuring the impact.

00:08:36: The old, sweet versus best-of-breed debate, which David Chan noted is still going on,

00:08:41: where are we now?

00:08:42: Well, Gen AI seems to be tilting the scales towards composable.

00:08:46: Franz Reversma explained that Gen AI has changed how we think about content.

00:08:51: It puts content right at the center, managed within a composable ecosystem.

00:08:55: He described it as being like Lego for content.

00:08:56: Lego for content.

00:08:57: I like that.

00:08:58: Modular, reusable pieces.

00:09:00: Exactly.

00:09:00: Designed for faster experimentation, more targeted assembly.

00:09:03: Sounds good in theory, but we need proof, right?

00:09:06: Does simplifying the stack, maybe using this composable approach, actually deliver measurable ROI?

00:09:11: Are there examples?

00:09:12: Yes, there are.

00:09:13: And they really back up this whole pragmatic, reset idea.

00:09:17: Tobias Hartung shared a great case study about meta.

00:09:20: Meta,

00:09:20: what did they do?

00:09:21: They used Salesforce Data Cloud to consolidate their MarTech stack.

00:09:25: Went from six tools down to just three.

00:09:27: Okay, and the results?

00:09:29: Pretty dramatic.

00:09:30: Massive reduction in operating expenses, obviously.

00:09:34: But also, they increased campaign velocity.

00:09:36: launching over thirty more campaigns per quarter.

00:09:39: And crucially, it unlocked net new revenue because it could finally target the mid-market effectively.

00:09:44: Wow.

00:09:44: Fewer tools, more campaigns, more revenue.

00:09:47: That's the dream scenario for Stack Hygiene, isn't it?

00:09:49: It really is.

00:09:49: It shows the payoff.

00:09:50: But

00:09:51: if successes like that are possible, why do we still hear constantly that measuring MarTech ROI is such a struggle?

00:09:58: It feels like a disconnect.

00:09:59: It is a huge disconnect.

00:10:01: And that brings us right back to the core issue.

00:10:03: McKinsey released some data recently.

00:10:05: They interviewed fifty senior leaders at Fortune-Five Hundred companies.

00:10:08: Yeah.

00:10:09: And not one of them could clearly articulate the ROI of their martech investment.

00:10:13: Not

00:10:13: one, seriously.

00:10:14: Not one.

00:10:15: They were tracking operational stuff, you know, how many emails sent, open rates, things like that, but not the metrics that actually tied to business value, like revenue growth or customer lifetime value.

00:10:25: So yeah, even with cleaner stacks, potentially on the horizon, the measurement battle is definitely still being fought.

00:10:31: So boiling it all down from these last couple of weeks on LinkedIn, What's the big picture?

00:10:36: The main takeaway for B to B marketers listening.

00:10:38: I think it's this.

00:10:40: Strategic B to B marketing is really going through a much needed pragmatic phase.

00:10:44: It's about cleaning up your tech stack, empowering your MOPs team to handle AI governance, locking down your first party data strategy, and fundamentally rethinking AI.

00:10:55: Stop seeing it as just another tool.

00:10:58: Start seeing it as potentially the central orchestrator for how you operate.

00:11:01: get the foundation solid, then leverage AI intelligently.

00:11:05: Pretty much.

00:11:05: But there's a flip side to this AI agent discussion, a future looking question for you to think about.

00:11:11: We've talked a lot about AI agents working for marketing teams.

00:11:14: But what happens when your customers start using their own sophisticated AI agents?

00:11:19: Aaron Chagrin posed this challenge.

00:11:21: If a customer's personal AI handles, say, ninety-five percent of their buying journey, researching, comparing, maybe even negotiating, how do you build brand loyalty?

00:11:30: How do you create delight?

00:11:32: How do you form any kind of relationship with an algorithm?

00:11:34: Huh.

00:11:35: That's a really interesting question.

00:11:36: Building brand with a bot.

00:11:38: Exactly.

00:11:39: That's the next strategic mountain to climb, I think.

00:11:41: Definitely something for you to mull over.

00:11:44: Okay, that wraps up this discussion.

00:11:46: If you enjoyed this deep dive, new additions drop every two weeks.

00:11:49: Also check out our other additions on account-based marketing, field marketing, channel marketing, AI in B to B marketing, go-to-market, and social selling.

00:11:57: Lots to explore there.

00:11:58: Absolutely.

00:11:59: Thanks for listening and don't forget to subscribe for more deep dives.