“You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”
This is the most important thing I've ever written. If you don’t have time to read it now, bookmark it for later.
Every few decades or so, a function of marketing goes through a structural shift. The old playbook dies, a new one is born, and the teams that identify it early reap all the benefits. Right now, lifecycle marketing is at one of those inflection moments.
Quick note: I've been working on something behind the scenes for the last few weeks that I’m super excited about. On July 16th, I'm hosting our first-ever virtual workshop at 12PM CST. I promise this is not something you want to miss.
Now back to the topic of codification. 😁
Every industry is feeling the pressure to adopt AI. And everyone is swimming as fast as they can. But almost nobody is sure whether or not they’re heading in the right direction, and only a few are talking about how overwhelming it all is.
Currently, three types of lifecycle marketing teams are forming.
Team 1: Has never or rarely used AI
Team 2: Uses it to speed up their workflows and processes
Team 3: Is completely rethinking everything around AI
The lifecycle marketing teams of the future won’t look anything like the lifecycle teams of the past. This post is about Team 3, and why I'm betting on something I’ve come to define as codified marketing: the act of making your proprietary marketing knowledge accessible to AI agents wherever it already lives, so they can operate on it directly.
It’s the best term I can come up with to define the core problem we’ve heard from hundreds of CMOs and lifecycle teams here at Hightouch, as well as countless conversations I’ve had myself.
The pattern hiding in plain sight
The fundamental job of lifecycle marketing has and will always be the same: deliver the right message to the right person, through the right channels, at the right time.
You’d think the majority of the time for a lifecycle marketer would be spent launching campaigns and running experiments to capture learnings to improve those campaigns. The unfortunate reality is that their time is often spent on operations rather than big strategic bets that could move the needle.
And because lifecycle knowledge is often scattered across different tools and locked in the head of whoever’s been there the longest, every new campaign often starts from zero with someone trying to remember what they already know.
The first wave of AI didn't change this; it just sped up certain parts of the traditional workflow: more copy variations, better subject lines, faster audience builds, faster reporting, and better QA. These are meaningful singular points of improvement. But everything still feels manual because it is manual. Speeding up individual steps doesn’t change the shape of the work itself.
As a result, the gap between what teams could ship and what actually gets shipped grows bigger and bigger every quarter, and ambition rarely ever intersects with reality.
What it’s like being a lifecycle marketer
The same underlying shift that rewired how software gets built is now happening in marketing. And the current model we’ve operated under for years — the briefs, the agencies, endless cycles of approvals, the data tickets, the handoffs, the QA tests, the design reviews, the legal reviews — is being rewritten because for the first time, you can now point AI toward those seemingly intractable problems.
“AI transformed software engineering because every repo has years of structured data: files, code, tests, history, docs, and every decision the team ever made, all unified and structured in one place. Once AI could reason across all of it, software engineering fundamentally changed forever. That same transformation is taking place in marketing right now.”
What the new world looks like
Getting the full value from AI requires more than prompt engineering. AI needs access to the proprietary context behind every good marketing decision: who the customer is, what they’ve bought, how they’ve engaged, what has worked before, how your brand communicates, and the legal and operational guardrails that govern what can go live.
When AI agents can operate across that context, the role of the lifecycle marketer changes. Instead of manually coordinating every campaign, marketers can build and manage agent systems that continuously decide what to create, who should receive it, when it should be delivered, and what to do next based on the results.
Most companies already have the context required to make this possible. The problem is that it is scattered across systems. Customer and behavioral data live in the warehouse. Brand assets live in a DAM. Creative lives in Figma. Performance data lives across ad platforms and engagement tools. Brand guidelines, legal requirements, and institutional knowledge are buried in documents or held in the heads of experienced employees.
The teams pulling ahead are connecting three layers of the lifecycle marketing system:
Context: Everything that makes your customers, brand, and business unique.
Content: The creative, messaging, and experiences you deliver.
Orchestration: The systems and workflows that bring campaigns to market.
The new mental model for lifecycle marketing
Most teams already have pieces of all three. What they lack is a connected context layer that allows AI to reason across them. We’re going to go much deeper on what this looks like in practice in the upcoming webinar, including how teams are actually building it, but this framework is a useful place to start diagnosing your program.
Codification does not require moving everything into a new platform. Your warehouse remains your warehouse. Your DAM, ESP, ad platforms, creative tools, and documents remain where they are. The goal is to make the knowledge across those systems accessible to AI through a shared, composable context layer.
The warehouse should anchor this framework because most lifecycle decisions begin with the customer. It contains the behavioral history, purchases, product usage, and other signals that provide the most complete view of who someone is and what they may need next. The other systems add the brand, creative, and operational context agents need to act on that understanding and turn it into a customer experience.
The biggest callout here is that context does not need to live in one place. It needs to be accessible through one connected layer without replacing the systems that already run the business or creating another bottleneck. As agents use that context, execute campaigns, and write the results back into the existing stack, each campaign makes the system smarter and more useful.
AI isn’t just accelerating the lifecycle workflow. It is reducing the distance between imagination and execution. As that distance shrinks, marketers can spend less time coordinating work and more time creating, experimenting, and learning. The opportunity is not simply greater efficiency. It is a fundamentally higher rate of execution and learning.
“AI isn't just accelerating the lifecycle workflow. It's reducing the distance between imagination and execution. And when that distance shrinks, marketers spend less time coordinating work and more time building, creating, and experimenting. That's where the real opportunity lies, not just in efficiency, but in unlocking entirely new levels of velocity.”
The same campaign 2 ways
At this point, you’re probably thinking: this sounds great in theory, but nobody is actually working this way. That’s a fair reaction, and to some extent even true, as the majority of lifecycle teams aren’t quite this far along with AI. I’ve gotten a sneak peek at the ones that are though, and once you see what the new workflow looks like, it’s really hard to go back.
To make this concrete, I'm going to walk through a simple example:
How workflow changes when marketing teams own more than the campaign brief.
The old workflow was a series of handoffs strung together by the marketer, who spent most of their time translating the same idea into a new language for whoever needed it next.
The data team needed the segmentation criteria spelled out before they could build the audience.
The agency needed the brief on positioning, the campaign goals, and the creative direction.
Legal needed the constraints documented before approving anything that would actually go out.
Ops needed the timing, the cadence, and the suppression logic to coordinate the launch across email, SMS, and push.
That’s four teams and four conversations (and this is simplified: we’ve spoken with teams that take well over 300 steps to send an email).The campaign eventually ships, but only after weeks of back and forth, and almost always different than what the marketer originally had in mind.
The new workflow collapses all of that into a single handoff.
The marketer delegates the campaign to agents, which already have all the context the work depends on.
Agents draft the campaign, then the marketer reviews the result and refines it.
Agents coordinate the launch across the different channels with the right timing and cadence, with the right approvals.
“Most growth bottlenecks aren’t strategy problems—they’re execution drag. A marketer has an idea, but then spends days translating it across systems and teams. Codified lifecycle changes that dynamic. You can go from idea to live experience in hours instead of weeks, which fundamentally changes what a growth team is capable of.”
The compounding gap
This idea of codified marketing doesn’t just make the team faster in terms of step-change; it changes how impact compounds entirely. The teams adopting this new operating model are starting to compound in three directions at once:
Vertically, because each layer of the stack multiplies the value of the layer below it.
Horizontally, because each layer of the stack gets richer with use.
Exponentially, because volume creates insights to act on in future campaigns.
Linear growth vs. exponential growth
The exponential one is what most people miss, and it's the one that actually matters the most. A team that ships 100 campaigns with a 100% success rate will lose to a team that ships 1000 with a 30% success rate, because every one of those 1000 campaigns — the 300 that worked and the 700 that didn't — is a data point the system learns from and acts on next time.
Please note: I’m not at all saying that volume is the only thing that matters in marketing. However, I am saying that AI unlocks a depth of personalization and testing that just wasn’t feasible in the old world. And that unlock means campaigns for segments that would have previously been too small to justify and variants that would have never been prioritized before. Teams can now go ten levels deeper on personalization because the effort and cost to do so have dropped so dramatically.
This is why large technology companies often seem almost eerily good at personalization. It is not necessarily because they have better lifecycle teams, but rather because they’ve spent decades building proprietary systems that run learning loops at enormous scale. AI changes that equation. By operating across a codified context layer, lifecycle teams unlock a kind of continuous learning and compounding advantage that once required years of infrastructure investment.
How this change will impact your role
The job of lifecycle is undoubtedly changing, as teams shift away from building and launching campaigns from scratch and instead begin to manage agents for every distinct part of the traditional marketing workflow.
Agents need guidance, and they need to be driven by people who know how the work was done before AI and deeply care about the end-user — people who understand the nuances of emotion and the relational side of humans — people who don’t just view customers as numbers in a database where LTV simply moves up or down. The teams that win in this AI world will be the ones that adapt and reorganize around the things that humans do uniquely well and the things that machines do uniquely well.
And the reality of that is that some work will (and is already) compressing dramatically — audience definition, localization, execution, Q/A testing, HTML building, and many other things. But I also believe that new roles and opportunities will be born out of this because the work that codifies away is probably the work that nobody liked doing to begin with. Which means what’s left is the work you got into marketing to do: bring ideas to life and make customers happy.
One last thing…
No one knows the future, but this post is a synthesis of the countless conversations I’ve had as someone who is deeply embedded in lifecycle and martech, as well as what Hightouch has heard broadly at talking lifecycle teams in every industry.
As always, I welcome your feedback. And there are three ways you can do that: (1) comment on this post, (2) share it on LinkedIn and tag me, (3) forward this to someone else.
— Luke