An unspoken law of inertia often governs marketing teams in older organizations:
“We’ve always done it this way.”
But ask someone why, and the room suddenly gets very quiet.
That workflow? No one remembers who created it. That monthly report? No one has read it since the last Twilight movie hit theaters, but someone keeps sending it.
Somewhere along the line, the rationale (and the people) behind these decisions vanished, but the rituals stuck around.
Legacy, baby.
Now layer in a few strategic reorgs, tack on some martech, and you’ve got a marketing machine that runs (efficiently, even), but nobody can explain how it works or why it exists.
The problem with that lack of understanding is that when people try to change processes, no one’s sure which ones are holding everything up. And everyone’s afraid of sawing through a support beam.
Maybe that six-step review process exists for a reason. Maybe that fusty old channel is still doing something worthwhile.
Every change feels risky because the current stack may be inefficient, but it hasn’t collapsed. Yet.
Typically, this inherited confusion is the stuff of hallway grumbles. They’re harmless inefficiencies that might lead to a slightly more complex technology implementation process. The kind of thing you patch as you go.
But the pressure to view AI as a potential replacement for marketing team members introduces a new twist. Now, teams aren’t just losing institutional memory over time; they’re actively replacing it.
When AI tools stand in for humans, they quietly erase the institutional memory of why things were built the way they were.
Here’s the kicker: AI doesn’t inherit wisdom, and it doesn’t arrive with any pre-installed. It has to be taught.
Yet, when no one’s left to explain the why behind the what, there’s no teacher in the room. So, sure, the “early adopter” senior exec gets what they want: The machine keeps running, faster than ever.
But it’s learning from a system no one understands. The forgetting isn’t just continuing, it’s accelerating.
Erasing institutional memories
If there’s any foundational principle that surpasses speed in modern business culture, it’s efficiency. And AI is emerging as the shiny new rationale for reorganizing marketing and shedding headcount for efficiency’s sake.
“We don’t need the team. We’ve got tools!” goes the logic.
Yes, tools do remember things. But they only remember things that were formally documented.
They remember the final slide deck, not the institutional wisdom gained in the (sometimes heated) whiteboard sessions behind it.
They store the final messaging framework from the agency, but not the emotional customer interviews the team conducted to shape it.
The strategic memory that lives between the lines — the judgment, the context, the why, the wisdom — gets lost. That’s the memory that fades when humans leave, and no machine can summon it back.
When you replace people with technology (any technology), you’re not just automating tasks. You’re also quietly erasing institutional memory.
There’s value in the swap. Efficiency. Scale. Maybe even effectiveness. But there’s always a cost, too, that’s usually hidden and almost always delayed.
AI might crank out tomorrow’s blog post, but it doesn’t know why the blog exists. It doesn’t remember that a heated debate in an offsite strategy meeting sparked the post. It doesn’t care that the brand team once banned the word “innovative” from headlines out of sheer fatigue. (Spoiler: Overuse stole all its meaning.)
Marketing operations in the dark
Forgetting isn’t just an AI problem. It happens even without the robots.
Strategies outlive their creators, rituals survive without reason, and no one can quite explain how the machine still runs.
Three CMOs ago, some team set a content or brand direction that made perfect sense at the time. Then they left. Their deputies left. Even the intern who vaguely remembers why the blog went from daily to weekly to “whenever someone remembers” is long gone.
I see “zombie ops” (processes that aren’t quite alive but won’t die either) even in small organizations. Things just lurch forward, quietly devouring resources and making it harder to do anything outside the “this is how we’ve always done it” box.
Maybe you’ve seen it in your company:
A 17-step approval workflow no one understands
Standards, such as a taxonomy, that haven’t changed since three acquisitions ago
A social media governance plan that was last updated when TikTok was still called Musical.ly
A weekly newsletter that exists because ... it always has?
Even core marketing elements start to become rote. What began as strategic differentiation starts to drift until the offering no longer matches the market, but no one wants to touch it.
The product reflects yesterday’s assumptions, not today’s needs. “Marketing legend says this feature was a game-changer in 2019” becomes the explanation, rather than a reason to evolve.
New teams follow the roadmap but misunderstand the intent. The result? A business that looks aligned on paper but feels out of step in practice.
Whatever the reason (turnover, tech, or time itself), a feeling of urgency about marketing’s institutional memory is growing.
But what can you do about it?
How to reset institutional memories
Until recently, inherited confusion was mostly a background annoyance — something you could fix in the margins. You patched what needed patching and carried on.
So, it’s tempting to think the answer is just as simple. Update the strategy. Swap out the broken parts. Rebuild the messaging. Refresh the content plan.
Easy, right?
But that approach only works when the problem is isolated to one channel, one team, or one tactic that needs a reset.
Today, the entire marketing engine is being “optimized” at once. Tech stacks, org charts, workflows, and brand narratives are all in flux. And in that kind of environment, surface-level updates don’t cut it.
It’s marketing gentrification: Slap some shiny new processes on the facade and call it a transformation, when what you need now is a structural reset.
Here are my suggestions on how to approach one:
Audit like an archaeologist
Take on the important (and, yes, boring) initiative to treat your current ops and brand materials like ancient ruins.
What’s here? Why might it have existed? What layers were added later? Document everything, even if it’s nonsense.
Figure out what still works
Don’t assume everything’s bad. Some rituals may have real value. That old webinar series? Still pulling leads. That brand voice? Surprisingly resonant.
Keep what works, question everything else.
Talk to the ancestors (if you can)
Reach out to former employees or consultants. Ask what constraints or ideas shaped the original strategy. Try to reassemble the initial wisdom of the day.
It’s like assembling original versions of ancient texts. You can conceive better new ideas if you understand why the old ideas made sense.
If that’s not an option, look to industry benchmarks to rebuild missing context.
Design and evangelize a new strategy that the existing team understands
Craft something new that your current team understands and believes in.
Write it down. Explain it. Make sure every piece has a clear “why” behind it.
Then tell the story of the strategy. This is the important part.
Don’t just unveil a PowerPoint and call it a day. Roll out the new approach with context and conviction. Treat it like a rebrand: People need to feel the purpose behind the shift.
Phase it in and let it breathe
You don’t need to launch the new system all at once. Pilot it. Tweak it. Make space for feedback.
The goal is to co-create something sustainable, not dictate it from above. This process doesn’t have to be anchored in speed.
It’s institutional change. It’s meant to stick.
Institutionalize the wisdom
Create living documents, not stone tablets. Make knowledge sharing a habit, not a one-time event.
Use mentorship, rituals, post-mortems, and culture to make the strategy sticky.
AI can’t remember what it never knew
As you begin your AI journey (whether it’s a forced march, an eagerly awaited adventure, or something in between), take time to ask what's more important: optimizing for output or preserving what makes us coherent?
Slow down. Ask hard questions. Invite your team to dig through the rubble with you.
Because the only thing worse than repeating the past is repeating it faster.
It’s your story. Tell it well.
Subscribe to workday or weekly CMI emails to get Rose-Colored Glasses in your inbox each week.
HANDPICKED RELATED CONTENT: Fast, Broken, and Stuck? Your Content Team Needs Processes To Break