Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

AI is everywhere, but marketing leaders face a harder question: Where should we actually use it?

In this episode of B2B Orchestration, Ian and AJ unpack a practical approach to bringing AI into a marketing organization. The conversation starts with the human side of adoption, including the natural anxiety people have about AI changing their jobs, and argues for positioning AI as a way to enhance and supplement the work teams already do.

From there, the discussion gets tactical: start with technology you already own, identify concrete use cases, create space for experimentation, establish appropriate governance, and focus investment on measurable outcomes instead of chasing every new AI tool. The conversation even uses Adobe Marketo as an example of looking for AI capabilities inside technology an organization is already paying for.

The episode ultimately explores the question every organization will face: when should you buy an AI capability, and when does it actually make sense to build?

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Marketers aren’t lying to executives — they’re just explaining the end result without the how. That’s the distinction this conversation is built around, and it’s why multi-touch attribution keeps losing credibility with CFOs. The theory is elegant: distribute opportunity value across every touchpoint that contributed to a deal. But once you ask how the model actually assigns that credit, the story falls apart — it’s glorified last-touch dressed up as science, and the moment a touchpoint doesn’t get “the credit it deserves,” someone manually adjusts the model. Add in the fact that no one can closed-loop report an LLM citation, and you’ve got a measurement framework that’s been running on borrowed trust for years.

The conversation traces where that trust erodes and what’s replacing it. Instead of assigning dollar values to individual milestones — a blog visit, a badge scan, a demo — the shift is toward understanding causation: what actually moved someone to engage, and can you replicate it. This is where journey analytics and Markov modeling earn their keep — not by pricing out each interaction, but by modeling the path itself as the unit of value. That reframe matters more than it sounds: it’s the difference between measuring activity and measuring intent.

The second half applies the same lens to lead scoring. The 100-point threshold gets called out for what it is — a playbook that expired a decade ago, when scoring was never meant to declare someone “ready to buy.” It was designed to indicate engagement and content affinity, full stop. The episode walks through how that got corrupted — MQL, then AQL (automation qualified lead) as a stopgap after sales stopped trusting MQLs, then back to a form of MQL — and lands on where scoring is genuinely useful today: as a decay-based signal of fading intent, not a qualification gate.

The takeaway for anyone running attribution or lead scoring today: stop defending the model and start defending the story it’s supposed to tell. If you can’t explain how a number got generated, the number won’t survive contact with a skeptical buying committee — whether that committee is your CFO or your sales team.

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

This episode of the B2B Orchestration Podcast tackles a question most MarTech leaders are avoiding: if AI agents can already write, approve, and deploy campaign content for pennies, what happens to the traditional budget-and-roadmap model that’s run B2B marketing ops for a decade?

The conversation opens with a blunt read on disruption speed — the host points to Chegg and Quizlet as early casualties of AI eating search-dependent business models, then pivots to what that pattern means inside marketing operations. The guest, who runs an AI consulting practice, walks through a real deployment: an agent built on Claude (Sonnet for writing, Opus for technical work, Fable for planning) that takes a content brief, writes the copy, and pushes finished emails directly into Marketo or Adobe Journey Optimizer via API — collapsing what used to be a full role (or an outside agency engagement) into a workflow a marketer just approves and sends. The number that anchors the whole episode: a 10-email nurture sequence, once the template and brief exist, costs roughly $0.10 to generate.

That economic shift forces a second conversation — how do you justify budget for something that cheap, and how do you measure value when the unit economics don’t look anything like the old model? The guest argues this is actually an easier budget conversation, not a harder one: leaders don’t approve budget for vague transformation anymore, they approve budget for a visible, ownable outcome. “I can see this thing, I know what it’s going to do” beats a slide deck every time.

The back half goes further, into why the standard 90/180-day or 12-month transformation roadmap is becoming a dead pitch. With model capability shifting quarterly — new releases, new agentic tools, new price points — multi-year MarTech commitments are a hard sell to execs who genuinely don’t know what next quarter’s tooling will make possible. The episode closes on a speculative but pointed thread: whether GitHub becomes an “Etsy for skills,” where individual builders sell narrow AI workflows directly, and why enterprise buyers — compliance, security review, and vetting requirements being what they are — are structurally unlikely to ever buy that way, even as solopreneurs and small shops already are.

Takeaway for MarTech/RevOps leaders: stop selling multi-year transformation. Sell a visible, cheap, ownable outcome first, then use that proof point to open the budget conversation for what comes next.