In this blog post I'm going to walk you through the change management side of AI marketing implementations, the part where you've built the AI workflow and now your team has to use it. The version that addresses resistance, fear, and habit instead of pretending those don't exist.
Most "AI change management" content in 2026 is written by HR software vendors or enterprise transformation consultancies. It's designed for organisations restructuring around AI at thousand-employee scale. Useless if you're a marketing leader trying to get a team of 4-30 people to adopt new workflows.
This article addresses the smaller-scale, marketing-specific reality. The patterns I see in actual rollouts. The interventions that work. The ones that don't.
I've been a marketing consultant for twenty-one years. I went all in on AI in 2024. I've watched smart AI implementations die because nobody planned the team side. I've also watched mediocre AI implementations succeed because the team side was nailed. The team side determines more outcomes than the technical build.
By the end of this blog you'll know the specific resistance patterns to expect, the interventions that work, and the order to do them in.
TL;DR
Three things determine adoption:
- Whether the team understands what changes for THEM personally (most leaders skip this)
- Whether the team trusts the AI outputs (built through transparency, not assertion)
- Whether the workflow makes their job easier or harder (workflow design matters)
Five resistance patterns to expect:
- Fear of replacement (the loudest)
- Quality scepticism (the quietest, the most legitimate)
- Habit inertia (the most underestimated)
- Identity threat (the hardest to address)
- Tooling fatigue (the most preventable)
Six interventions in order:
- Communicate the why and the "what changes for you" before the build
- Involve the team in workflow design
- Pilot with self-selected early adopters
- Document the quality bar explicitly
- Deprecate replaced manual work openly
- Reward adoption, not just compliance
Why marketing teams resist AI specifically
Three reasons marketing resistance is different from other functions:
1. Marketing is identity-coded. Most marketers chose the field because of the creative-craft elements. AI threatens those specifically. Resistance often comes from people who'd otherwise be aligned.
2. Marketing outputs are visible. When sales operations AI makes a mistake, it lives in the CRM. When marketing AI makes a mistake, it's published. The stakes of bad AI output feel higher.
3. Marketing has been burned by past tech promises. "Marketing automation will revolutionise everything" was the 2010s pitch. Many marketers remember the gap between promise and delivery. They're cynical about new tech for legitimate historical reasons.
The interventions below address all three.
The 5 resistance patterns
Pattern 1: Fear of replacement
The signal: "Are you trying to replace us with AI?"
The reality: This is the loudest pattern but usually not the most active resistance. Most fears of replacement aren't about NOW, they're about 2-3 years from now.
The intervention: Address it explicitly and early. Don't dance around it. State clearly: "We're building AI workflows to remove [specific operational work]. Your role changes to focus on [specific higher-use work]. The team size for this function isn't changing in 2026."
If your team size IS changing, say that too. Lying or hedging creates worse resistance than honesty.
Pattern 2: Quality scepticism
The signal: "I tried ChatGPT for this last year. The output was rubbish."
The reality: Most marketers have tried consumer-grade AI on a marketing task and gotten generic output. Their scepticism is empirical, not irrational.
The intervention: Show specific examples of AI outputs from your custom workflow vs the generic ChatGPT comparison. Include side-by-side. Acknowledge the legitimate scepticism. Demonstrate the difference structured workflows make.
Avoid: dismissing past experiences, calling them "wrong," or insisting AI has gotten better.
Pattern 3: Habit inertia
The signal: Six weeks post-launch, the team is still doing things the old way.
The reality: Habits are stronger than logic. Even when AI workflows save time, the manual process is what the team is comfortable with. The friction of switching outweighs the benefit per task in the short term.
The intervention: Three moves:
- Block calendar time for the team to use the new workflow in the first 4 weeks (forced repetition)
- Make the old manual process slightly harder (e.g., remove the spreadsheet template from the shared folder)
- Pair team members so they do the new workflow together initially
Pattern 4: Identity threat
The signal: Quieter resistance from senior team members. Subtle obstruction. Phrases like "the AI doesn't get what we do."
The reality: Senior team members have built careers on specific craft. If AI does that craft, their identity is destabilised. This is the hardest pattern to address because it's not really about AI.
The intervention: Public elevation of the human-only craft elements. Make it explicit what AI doesn't do (judgement, taste, strategy, relationships) and that those become MORE valuable, not less. Re-define seniority around those elements.
Don't: try to convince them AI is "just a tool", that minimises their concern.
Pattern 5: Tooling fatigue
The signal: "Another new tool? We just learned [previous tool]. I can't deal with another one."
The reality: Legitimate. Marketing teams typically use 8-15 tools. Each new one adds cognitive load.
The intervention: Two moves:
- Aggressively deprecate replaced tools. New AI workflow ships, old tool license cancelled within 30 days. Make the addition feel like net subtraction.
- Bundle adoption training with existing tool training, not as a separate burden.
Six interventions in order
Intervention 1: Communicate the "what changes for you" before the build
Most leaders skip this. They announce "we're implementing AI workflows for sales operations" and assume the team will read about specifics later. The team interprets the absence of detail as worst-case.
What to do instead: Before the build kicks off, hold a 60-minute team session covering specifically:
- What workflows are changing
- What stays the same
- What each person's day will look like differently in 90 days
- What worry you're hearing from them is founded vs not
Timing: Day -7 before kickoff. Not day 1. The week before matters.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Intervention 2: Involve the team in workflow design
The team that designs the workflow uses the workflow. The team handed a workflow resists it.
What to do: Make sure 1-2 team members are in the workflow design sessions with you and the consultant (or in your build sessions if internal). Not as observers, as contributors. Their fingerprints on the design create ownership.
Don't: skip this to "save time." It costs you 2-3 days upfront and saves 6-12 weeks of adoption struggle.
Intervention 3: Pilot with self-selected early adopters
Every team has 2-3 people who'd like AI. Find them.
What to do: Open early access. Let early adopters use the workflow in production for 4-6 weeks before everyone else. They become internal champions. They surface issues quietly without making the broader team panic.
Don't: force everyone to use it on day one. The early-adopter cohort de-risks the broader rollout.
Intervention 4: Document the quality bar explicitly
If "good AI output" is subjective, every team member sets their own bar. Some accept anything, some reject everything.
What to do: Document specifically what "ship-worthy" output looks like vs "needs editing" vs "reject." With examples. Get senior sign-off on the criteria.
This protects everyone: the people who'd accept too much, and the people who'd reject too much.
Intervention 5: Deprecate replaced manual work openly
The biggest sign that AI implementation works: things STOP happening, not just start.
What to do: Publicly list what manual work is no longer happening. Cancel the Tuesday morning report meeting if AI is producing the report. Take the manual spreadsheet template out of circulation. Make the deprecation visible.
Don't: leave both running indefinitely. Two parallel processes create confusion and resentment.
Intervention 6: Reward adoption, not just compliance
If the only signal you send is "use the new workflow or you're not on board," you'll get compliance theatre. People will go through the motions while quietly reverting.
What to do: Recognise and reward genuine adoption. Specific praise for team members who improved the workflow. Public credit when AI-assisted work outperforms the previous baseline. Make adoption feel like winning, not just complying.
Don't: punish slow adopters in the first 60 days. Most people take that long to internalise new workflows.
The 90-day adoption checklist
| Day | Milestone |
|---|---|
| Day -7 | "What changes for you" session held |
| Day -3 | Team members in workflow design have signed off |
| Day 1 | Build kicks off; weekly team updates begin |
| Day 14 | Early adopters identified and given access |
| Day 21 | Quality bar documented and signed off |
| Day 30 | First training session with full team |
| Day 45 | Production rollout begins |
| Day 60 | Manual process deprecation visible |
| Day 75 | First adoption review (who's using, who isn't, why) |
| Day 90 | Adoption metrics + retrospective + plan for next workflow |
If 30 days post-rollout the team isn't using the workflow, the change management failed regardless of how good the AI workflow is.
What if the team fundamentally won't adopt
Three possible reasons:
Reason 1: The workflow isn't better. The AI output is worse or the workflow adds friction the team is right to resist. Solution: re-evaluate the workflow, not the team.
Reason 2: Senior leadership undermined adoption. The team mirrors leadership. If a senior leader rolls their eyes at AI in meetings, the team will too. Solution: leadership alignment first.
Reason 3: You're trying to AI-replace work that's valued. Some of the work AI could automate is the work team members find meaningful. Removing it removes meaning. Solution: don't automate that work even if you could.
Don't reach for "the team is resistant to change" as the explanation. It's usually the symptom, not the cause.
Frequently asked questions
How small a team is too small for formal change management? Three people or fewer can usually wing it. Four or more needs structure. Marketing teams of 8+ definitely need the full process.
Should HR be involved in marketing AI rollouts? Generally no. AI implementation in a function is the function leader's responsibility. HR involvement often slows things down.
What if my team includes contractors and freelancers? Address them too. They're often the most resistant because their value proposition is most threatened.
How long should the full adoption process take? 90 days minimum for a single workflow. 6 months for a multi-workflow programme.
Can I outsource change management to my AI consultant? Partially. They can support but the leadership signal has to come from the function leader. Outsourcing the leadership part doesn't work.
Should I share metrics with the team during rollout? Yes. Transparent measurement reduces speculation. Even when results are mixed, knowing the numbers beats not knowing.
What if HR or legal demands input on AI rollouts? Engage them but don't let process gates kill momentum. Define their role narrowly (compliance review, communications input) and proceed.
Want help running the adoption?
I'm Lilach Bullock. I've been a marketing consultant for twenty-one years. I went all in on AI in 2024. I work with founders and marketing leaders who want AI to move their numbers, not just their tool stack.
If you want the full breakdown, here is everything I know about AI marketing.
