By Steve Raju

For Marketing and Advertising

Cognitive Sovereignty Checklist for Marketing and Advertising

About 20 minutes Last reviewed March 2026

AI tools like ChatGPT and Midjourney can generate dozens of campaign ideas in minutes. But that speed often masks a deeper problem: your team stops asking why something will matter to real people, and starts asking what the AI will approve. The craft knowledge that built memorable brands gets replaced by prompt engineering. This checklist helps you keep human judgement at the centre of your work.

Tool names in this checklist are examples. If you use different software, the same principle applies. Check what is relevant to your workflow, mark what is not applicable, and ignore the rest.
Cognitive sovereignty insight for Marketing and Advertising: a typographic card from Steve Raju

These are suggestions. Take what fits, leave the rest.

Download printable PDF
0 / 19 applicable

Tap once to check, again to mark N/A, again to reset.

Protect Your Strategic Thinking

Write your brand strategy before opening any AI toolbeginner
Your account team should document what the brand actually stands for, who cares about it, and why they care. This becomes your filter. AI tools will generate options that sound strategic but contradict your real strategy.
Name the specific behaviour change you want from your audiencebeginner
Not engagement. Not awareness. Real behaviour: purchase frequency, trial, switching from a competitor, or repeat visit. AI tends to optimise for volume metrics instead of the behaviour that matters to your client.
Write the one thing you need the audience to believeintermediate
Before briefing AI, articulate the single belief that would make your audience change their behaviour. This becomes your guard rail. AI will generate many things your audience might find interesting. Most will be irrelevant to this belief.
Define which audience segments need different reasons to careintermediate
Your brand does not speak to one audience. You might need to tell parents one thing about a product and teenagers another. Map these segments and the distinct reason each would care before AI writes audience copy.
List what has worked before and why you think it workedadvanced
Review your previous campaigns that moved the needle. What did they have in common? What was the insight about your audience? AI cannot learn this from your data alone. You have to teach it.
Set a clear constraint on brand tone and styleintermediate
Do not let AI choose your tone. A global brand might need to stay professional. A youth brand might need to feel chaotic. A luxury brand must avoid being too friendly. Write this down and test every AI output against it.

Keep Judgement in Your Creative Process

Generate only three to five campaign directions from AI, then stopbeginner
More options do not equal better work. They breed paralysis and cost you time. Ask AI for three distinct angles. Then use human judgement to pick the most defensible one. Evaluate it hard before going to production.
Spend a full day on why your chosen idea will work before you brief productionintermediate
This is where your strategic thinking becomes real. Write out the insight about your audience. Explain the behaviour change you expect. Show why the creative approach addresses that insight. If you cannot make this case, go back to step one.
Test your AI-generated creative against competitor campaignsbeginner
Pull five recent campaigns from your competitor set. Are you saying something different? Does your tone stand apart? If your AI output looks like a blend of what competitors are doing, it will disappear in the channel.
Have someone who knows your client brand review every piece before deliverybeginner
This is not a rubber stamp. This person should be able to say no. They should know the brand strategy, the audience, and what has worked. They catch the moment AI has generated something competent but off-brand.
Ask AI to write three reasons why its own creative will failintermediate
This reversal often reveals gaps that your team missed. The AI might expose an assumption you made or a detail about your audience you overlooked. Use its critique as input, not as gospel.
Assign one person to defend the creative choice and one to attack itadvanced
Before you lock in direction, have them spend an hour making their cases. The person defending finds the insight that makes it work. The person attacking finds the risk. This mimics the natural tension that produces good work.
Create an audit trail showing which AI outputs your team rejected and whyadvanced
This becomes evidence that you are making judgements, not just following AI suggestions. When a campaign underperforms, you can see whether the problem was the strategy or the execution of that strategy.

Preserve Insight and Craft Knowledge

Train your junior staff on how to read your audience insight before they use AIbeginner
Junior team members often go straight to the AI without reading the research. They miss the human truth embedded in consumer interviews and data. That truth is what makes an idea distinctive, not an AI feature.
Document the reasoning behind every significant campaign decisionintermediate
When you decide on a tone, a message angle, or a creative approach, write down why. This becomes institutional knowledge. It prevents the next person from asking an AI to solve problems that were already solved.
Run a monthly review of which metrics matter most to each clientintermediate
Media teams often optimise for click-through rates or cost per impression. But your client cares about sales lift or brand recall. Document what each client actually measures success by, then protect that metric from AI over-optimisation.
Preserve the account lead role as a strategic thinker, not a prompt writeradvanced
If your account leaders spend all day writing prompts and reviewing variants, you have lost them as strategists. Protect time for them to think about brand positioning, audience behaviour, and long-term client relationships.
Conduct a quarterly audit of your brand distinctiveness across all channelsadvanced
Gather your last three months of paid search copy, social media posts, display creative, and email campaigns. Could a competitor have written any of it? If the answer is yes, your AI outputs are averaging your brand.
Build a private database of what has worked for each brand you serviceadvanced
Do not rely on general AI training data. Create a searchable archive of your successful campaigns, the insights behind them, and the results they produced. Use this to brief AI instead of relying on its base knowledge.

Five things worth remembering

Related reads


Common questions

Should marketing and advertisings write your brand strategy before opening any ai tool?

Your account team should document what the brand actually stands for, who cares about it, and why they care. This becomes your filter. AI tools will generate options that sound strategic but contradict your real strategy.

Should marketing and advertisings name the specific behaviour change you want from your audience?

Not engagement. Not awareness. Real behaviour: purchase frequency, trial, switching from a competitor, or repeat visit. AI tends to optimise for volume metrics instead of the behaviour that matters to your client.

Should marketing and advertisings write the one thing you need the audience to believe?

Before briefing AI, articulate the single belief that would make your audience change their behaviour. This becomes your guard rail. AI will generate many things your audience might find interesting. Most will be irrelevant to this belief.

The Book — Out Now

Cognitive Sovereignty: How To Think For Yourself When AI Thinks For You

Read the first chapter free.

No spam. Unsubscribe anytime.