|
Listen to this article
Getting your Trinity Audio player ready...
|
-AI for Leaders
BLUF: Five AI prompts for marketing leaders, each built with Google’s TCREI framework for real work: a creative brief, competitive intel, a board summary, a difficult conversation and a strategy stress-test.
Topic: Generative AI · Productivity · Leadership
Most marketing leaders aren’t bad at AI. They’re bad at giving instructions.
You’d never send your agency a one-line brief. You’d get one-line work back. Yet that’s exactly what you send AI, several times a day.
These are five AI prompts for marketing leaders, each built with Google’s TCREI framework (Task, Context, References, Evaluate, Iterate) for work you actually do. Not the theory. The prompts. If you want the definition first, start with What Is the TCREI Framework? This is the version with the sleeves rolled up.
Research Spotlight
According to Harvard Business School’s Working Knowledge (Bojinov & McFowland, March 2026), AI helps people with adjacent skills take on unfamiliar tasks, but hits a wall when they lack domain expertise. It amplifies what you already know.
Each example below shows the one-liner you’d normally send, the full TCREI version, and the part worth stealing.
1. The AI prompt for a creative brief your agency can’t misread
What you’d normally send:
“Write a creative brief for our spring campaign.”
The TCREI version of what you should send:
Task: “You’re a strategy director who has read 500 briefs and rejected most of them. Write a one-page creative brief with a single-minded proposition, one audience and one action we want them to take.”
Context: “The campaign targets [audience]. The business goal is [goal]. Budget is [range]. Last spring’s campaign had strong reach but weak conversion, because the message tried to say three things at once.”
References: “Here is last year’s brief and the agency’s first concept. The concept was beautiful and missed the point. Here’s why: [one line].”
Evaluate (your job, not the AI’s):
Could two different agencies read this brief and come back with the same core idea? If not, it’s not done.
We will work with you to create a personalized plan to help you achieve your financial goals.
Iterate:
“The proposition is two ideas pretending to be one. Pick the stronger one and rebuild the brief around it.”
What you can steal: the Evaluate step is the same test you’d apply to a human-written brief. AI doesn’t lower your standard. It just gets you to the draft faster.
2. AI prompts for competitive intelligence that go past the obvious
What you’d normally send:
“What is [competitor]’s marketing strategy?”
The TCREI version of what you should send:
Task: “Act as a competitive intelligence analyst. Map how [competitor] positions itself: who it’s for, what it promises, and what proof it offers. Put it in a table, then a three-sentence summary.”
Context: “We position ourselves as [your positioning]. This analysis will inform [decision], so I care about where we overlap and where they’re exposed.”
References: “Below are their homepage copy, their last three press releases and five recent LinkedIn posts from their CMO.”
Evaluate (your job, not the AI’s):
Evaluate: Separate what they say from what they can prove. Flag any claim with no evidence behind it. And check every fact: AI fills gaps confidently, so anything it didn’t get from your references needs verifying.
Iterate:
“Now switch sides. If you were their CMO, where would you attack us first?”
What you can steal: the References step does the heavy lifting. Paste in the source material rather than asking AI what it remembers. Memory is where the confident nonsense lives.
3. Turning your data dump into a board-ready summary
What you’d normally send:
“Summarize this data.”
The TCREI version of what you should send:
Task: “You’re a chief of staff preparing a three-minute update for the board. Turn this data into one headline, three numbers that matter and one decision we need from them.”
Context: “The board cares about [revenue / reputation / risk]. They’ve asked twice why [metric] is flat. They don’t want the methodology. They want to know whether to worry.”
References: “Here’s last quarter’s summary. It landed well because it led with the answer.”
Evaluate (your job, not the AI’s):
Evaluate: Trace every number back to your source. AI misreads tables more often than you’d like, and the board will find the error you didn’t.
Iterate:
“Cut it to the three numbers that would change a decision. Drop anything that only proves we were busy.”
What you can steal: the Task names the decision, not just the format. That’s the difference between reporting data and offering insight, and it’s the half of the job AI can’t do for you.
4. Preparing for the difficult conversation you keep postponing
What you’d normally send:
“How do I give someone negative feedback?”
The TCREI version of what you should send:
Task: “Act as an executive coach. Help me plan a 20-minute conversation: an opening line, the two points I must land, and how to close with a clear next step.”
Context: “A strong team member’s work has slipped for two months. They’re respected and they know it. My goal is to reset expectations without losing them. They’ll likely say they’re overloaded, and they may be right.”
References: “Here’s our team’s feedback model and the expectations we agreed on at the start of the year.”
Evaluate (your job, not the AI’s):
Evaluate: Read the opening line out loud. If it doesn’t sound like you, it won’t work coming out of your mouth.
Iterate:
“Now play the team member. Push back the way they actually would, and don’t make it easy for me.”
What you can steal: the rehearsal. Iterate isn’t only for fixing output. It’s a safe place to practice before the real thing.
One rule: leave out names and anything identifying, and follow your organization’s AI policy. The prompt works just as well anonymized.
5. AI strategy prompts that stress-test your plan before your CFO does
What you’d normally send:
“Is this a good strategy?”
(It will say yes. It’s polite.)
The TCREI version of what you should send:
Task: “Act as two people: a skeptical CFO and the CMO of our biggest competitor. Give me the five hardest questions each would ask about this plan, ranked by how much damage they’d do.”
Context: “I’m presenting this to the executive team in two weeks. The plan assumes [key assumption, e.g. budget holds flat and our main channel keeps performing].”
References: “Here’s the strategy deck and last year’s results.”
Evaluate (your job, not the AI’s):
Evaluate: Sort the questions into three piles: answered, answerable, and “I hadn’t thought of that.” The third pile is the whole point.
Iterate:
“Which single assumption, if it turned out wrong, would break the whole plan? What would I see first if it were breaking?”
What you can steal: asking AI to argue against you rather than agree with you. You’re not looking for reassurance. You’re looking for the question you’d rather hear in your office than in the boardroom.
Research Spotlight
60% of marketing teams that adapted their measurement approach to AI report returns of 2–3× or higher.
Scale your marketing faster with AI.
Synthesized from 100 top YouTube videos +10 primary research sources. Includes stats, prompts, tool stack, and implementation roadmap.
✓ Easy to use
✓ Complete 15-Page Visual Playbook (PDF)
Frequently Asked Questions
The ones with your context in them. A template gets you started, but the value comes from what only you know: the audience, the stakes and what failed last time. Use the five above as starting structures, not finished prompts.
Yes. Keep the Task and References fixed, and rewrite the Context every time. Context is the part that changes with each project, and it’s the part people skip.
Two minutes, once you know the framework. The TCREI structure isn’t about writing more – it’s about thinking more clearly before you type. Most of the time, you already have the context in your head. The framework just stops you from leaving it there.
Not always. For simple tasks, “summarize this in three bullet points”, you don’t need all 5. But for any output you’d actually publish, present, or act on, yes. Running through the checklist mentally takes about 30 seconds and consistently improves the result.
Yes. The TCREI framework is model-agnostic. The underlying principle that structured input produces better output applies regardless of which tool you’re using. The specific outputs will vary between models, but the prompting approach is universal.
That’s what the Iterate step is for. A bad second output usually means your context was too vague or your task wasn’t specific enough. Go back, tighten those two elements first. If you’re still not getting there, add a reference – even a rough example of what good looks like is often enough to unlock a significantly better result.
Everyone who produces output that gets used externally. Research from Jasper’s 2026 State of AI in Marketing report found that 60% of teams with structured AI workflows report 2–3x returns but adoption of structured approaches remains low. The teams pulling ahead aren’t just using AI more. They’re using it more deliberately, at every level.
Models are getting better at handling vague input but the output quality ceiling still scales with input quality. A March 2026 Harvard Business School study found that AI amplifies existing expertise but can’t bridge a domain knowledge gap. In other words: the better you are at your craft, the more your prompting skill compounds. AI rewards the people who already know what good looks like.
What Is the TCREI Framework? explains the five steps and where the framework comes from. This post shows them at work on real marketing tasks. Start there if TCREI is new to you.
Follow your organization’s AI policy and use approved tools. When in doubt, anonymize. Most prompts work just as well without names or confidential figures.
Sources
- McKinsey & Company The State of AI in Early 2024 – mckinsey.com/the-state-of-ai-2024
- Google & Ipsos AI Works for America, 2024 – ipsos.com/googleipsos-ai-works-america-poll
- Google Career Certificates Prompting Essentials – grow.google/ai-professional-control
- Jasper State of AI in Marketing 2026 – jasper.ai/state-of-ai-marketing-2026
- Harvard Business School Working Knowledge, March 2026 – library.hbs.edu/gen-ai-boosts-productivity-but-cant-turn-novices-into-experts
Nicola Ziady is a marketing strategist helping leaders build visibility, authority, and strategy in the AI era.
Follow along on LinkedIn
AI citation strategy, marketing leadership, and the Invisibility Paradox – every week. 24,000 marketing leaders already there.
Follow Nicola on LinkedIn →By Nicola Ziady. Published: March 21, 2026. Last revised: October 18, 2026
