Why Marketing Orchestration Is The New Superpower
Jason Ing has a useful way of describing where most marketing teams are right now. "I used to joke that AI was like my intern," he says. "But now that intern is graduating into a teammate."
As the CMO of Typeface, Ing leads what he refers to as an enterprise-grade marketing orchestration engine. He has spent nearly 20 years at Microsoft, Amazon, and Gusto watching how marketing organizations absorb new technology, and the pattern keeps repeating where teams adopt a tool, get faster at producing individual assets, and then stop. He notes that they're treat AI as an accelerant for the work they were already doing, rather than a reason to rethink how the work gets done at all.
This gap between using AI and running marketing with AI — is where the real opportunity is. Most teams haven't crossed it yet.
Want more from The CMO?
Sign up for a free membership to complete reading this article:
Have an account? Log In
The Coordination Problem Nobody Talks About
Ask any senior marketer where campaigns slow down, and you'll hear the same answer that it's not in the creation, but in everything after it. Switching between the handoffs, approvals, version control, and sometimes even legal review.
Ing is precise about where the breakdown happens at scale. "The breakdown is not at the point of creation, but in coordinating the handoffs between the marketing team, the creative team, and IT. That's where speed, consistency, and quality gets lost."
AI-as-assistant doesn't solve this. You can use Claude to write a better email in half the time, but if that email still has to travel through four departments before it goes out, you haven't changed how fast your marketing moves. Marketing orchestration is the attempt to fix the coordination layer, not just the creation layer. In practice, that means building systems where a single strategic brief becomes the input for agents handling channel-specific execution, brand compliance checks, and variant testing — with humans owning the decisions that carry real strategic weight.
The results can be significant. Ing points to a concrete example from his own team, where he shares how "we used our Email Agent to personalize webinar promotion across distinct personas, and it drove 4x higher email-to-registration conversion and a 2x lift in click-through rate versus a generic blast, while taking minutes instead of hours to produce multiple on-brand variants."
Why IT Is Now In The Room
For most of the last decade, marketing technology operated at a comfortable distance from IT. Marketers made their own stack decisions and largely went about their business. This arrangement, and freedom marketers have is changing fast. Ing notes how "IT alignment is now a prerequisite before any work really starts," Ing says. "Once your AI system is looking at things like compliance, privacy, brand governance, and campaign data, IT becomes central. Companies that win with AI treat it as shared infrastructure, not a side project."
This shift runs deeper than technical integration. An orchestration system that pulls from your CRM, your content library, and your performance analytics is touching data that marketing doesn't own alone. As AI starts making autonomous decisions within those workflows, the stakes of a misconfiguration rise accordingly. Colleen Goepfert, a fractional CMO and executive advisor, has watched this dynamic surface a related problem, where "leaders are expecting transformation without governance. AI generates options for you, but it's weak at resolving trade-offs. Leaders still need to know what to choose."
The implication is that marketing can't treat orchestration as a technology project handed off to IT, any more than it can treat it as a marketing project that runs in isolation. The teams building durable systems are the ones where both sides own the outcome.
Clarity First, Speed Second
The failure mode for most enterprise AI rollouts isn't that the AI is bad. It's that teams move into automation before they've resolved the foundational questions, in what the brand stands for, what claims can and can't be made, what a good brief actually contains.
Ing has watched enough pilots fail to be specific about this. "If the inputs are vague, say an unclear brief, muddled positioning, missing proof, then AI cannot rescue it. The fastest wins came when we tightened the upstream work first, and then used AI to scale what was already directionally correct."
Jennifer Tomlinson, EVP of Marketing at QorusDocs, arrived at the same conclusion through a different route. "We had to release the assumption that adopting AI would automatically make our marketing smarter. What it showed instead was that AI amplifies whatever clarity or confusion already exists." The operational implication is that building an agentic marketing system starts with documentation work that looks nothing like AI adoption. For instance, codifying brand rules, approval paths, and the logic behind messaging decisions. That groundwork is what gives agents something reliable to work with.
Building Within What You Already Have
One of the persistent frictions in enterprise AI adoption is the existing tech stack. Most large organizations have spent years building systems that work, albeit imperfectly, expensively, but reliably, and introducing orchestration infrastructure on top of that is genuinely complicated.
Ing's approach at Typeface is to start with integration rather than replacement. He shares how, "we recognize that customers, especially large businesses, have long-term contracts with their CRM provider or their customer data platform. We're not requiring you to scrap everything. We fit within the way you're already doing work."
Andrea Tarrell, President of Tech Services at Trilliad, reinforces the case for patience here, suggesting that leaders "look to optimize what you have, then consider what's missing. And be honest about what work you can take on internally versus where you'd benefit from a partner. Your team had a full-time job before you started the migration." The teams that navigate this best aren't the ones with the most ambitious roadmaps. They move in smaller increments, document as they go, and resist the pressure to show dramatic transformation on a quarterly timeline.
Where Humans Stay
The architecture of any orchestration system is only as good as the decisions about what to automate and what to keep human. Automate too little and you've spent significant money for marginal gains. Automate too much and you're publishing content no one with actual judgment has reviewed.
Ing's delineation is specific, and he shares that "the calls that stay very human are positioning, narrative, and brand trade-offs, especially as it relates to what we want to be known for, what we are willing to say, and what we will not say. AI can bring options and evidence, but humans should make the call. That keeps accountability clear."
Tomlinson puts a sharper point on what AI still can't replicate in that space, highlighting how "AI struggles with restraint. It's very good at filling space and very bad at knowing when less is more. Knowing what to leave out, when to pause, and how something will land still requires human judgment." That editorial instinct, when to cut, when to simplify, when to say no to a tactic, comes from experience that doesn't exist in a training set. It comes from having watched something land badly and knowing why.
The Compounding Effect
For teams that want to move toward orchestration without committing to a full platform overhaul, the practical starting point is narrower than most expect. Pick one campaign type where coordination friction is highest. Map the current workflow, identify where things slow down, then find one handoff where an agent could handle the transfer more reliably than a human. Build that. Make it work. Document what good looks like.
That documentation becomes the foundation of everything else — the skills, the knowledge base, the compliance rules, the brand voice parameters. All of it feeds a system that gets more useful the more precisely it's instructed.
Ing's point about compounding is worth taking seriously. "If you improve through AI one percent every day, by the time you get to the end of the year, that really compounds — and you realize how much further ahead you are than if you hadn't started."
The teams making real progress on orchestration right now aren't the ones who declared an AI transformation. They're the ones who picked a specific coordination problem, fixed it properly, and moved to the next one.
What's Next
Ready to grow? Join us at The CMO Club to get insights, guides, and resources from marketing leaders right to your inbox.
