Trend-to-Brief AI Workflow
Turn emerging social trends into data-backed ad briefs in minutes.
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 6
- Confidence
- 98%
The workflow begins by searching a social platform for high-performing videos within a chosen niche and time window. Engagement metrics help identify candidates, but the system then moves beyond surface statistics by using multimodal AI to inspect each video's hook, message, proof, pain points, theme, and funnel stage. It also analyzes comments to uncover recurring questions, objections, and customer beliefs. These findings are combined with a brand bible containing the audience, voice, positioning, pain points, and preferred creative-brief format. AI synthesizes both sides into a data-backed brief that a marketer can edit and approve. The mechanism converts unstructured social evidence into structured creative direction while preserving human judgment at the final quality-control stage.
Origin
Mike Foutia developed the workflow for e-commerce brands and Facebook ad agencies that were manually searching TikTok for emerging hooks and angles. Extracted from Marketing Against The Grain.
Core principles
- 01Use proven organic engagement as an early signal for paid creative.
- 02Analyze the content and comments, not just headline metrics.
- 03Encode expert judgment into the analysis prompt.
- 04Combine trend evidence with brand and customer context.
- 05Treat AI output as an editable first draft rather than a final decision.
How to run it
- 1
Define the search target
Choose a niche keyword, date range, and result count. Keep the search focused enough that the returned content is relevant to the product and audience.
Pro tip Start with customer problems or desired outcomes rather than broad product categories.
Watch out A broad query can produce popular but commercially irrelevant videos.
- 2
Collect trend evidence
Use a scraper or platform API to retrieve videos, transcripts, comments, and engagement metrics. Rank or filter the results to identify promising creative signals.
Pro tip Search enough videos to distinguish a recurring pattern from a single viral anomaly.
Watch out Views and likes alone do not explain why a concept worked.
- 3
Analyze each candidate
Send selected videos to a multimodal model and extract the visual hook, message, proof, pain points, theme, and funnel stage. Customize the analysis fields to match how the team evaluates creative manually.
Pro tip Encode the questions used by the team's strongest strategist into the prompt.
Watch out Generic prompts will produce generic observations rather than useful creative intelligence.
- 4
Mine the comments
Analyze comments for questions, complaints, desired outcomes, objections, and repeated customer language. Use these signals to understand what viewers believe and what remains unresolved.
Pro tip Preserve recurring customer phrases because they can inform hooks and copy.
Watch out Do not treat every comment as representative customer evidence.
- 5
Add brand context
Provide the model with the brand voice, audience, pain points, positioning, campaign goals, and creative-brief template. Add historical winning-ad patterns when available.
Pro tip Maintain reusable context files so every brief starts with the same strategic foundation.
Watch out Weak or outdated brand context can make a strong trend produce an unsuitable brief.
- 6
Generate and review the brief
Synthesize the trend evidence and brand context into a structured creative brief. Have a marketer correct assumptions, sharpen the concept, and approve it for production.
Pro tip Ask the reviewer to focus on strategic differentiation rather than rewriting every sentence.
Watch out Do not send the first AI draft directly into production without human review.
In the wild
The demonstration searches TikTok for gut-health videos, retrieves engagement data, and analyzes a promising video's acne-skin opening, proof, theme, funnel stage, and comments. The system combines those findings with an AG1 brand bible and brief template, producing a campaign objective, target audience, pain points, and the message that AG1 can replace confusing DIY gut-health remedies.
→ The team moves from an observed TikTok trend to an editable advertising strategy in under 15 minutes.
A supplement brand searches videos about sleep problems and analyzes comments under high-performing competitor content. Repeated questions about morning grogginess become a new objection-led hook. The workflow combines that language with the brand's non-drowsy positioning and produces a creative brief for paid-social testing.
→ The brand gains a customer-language concept grounded in current market evidence.
Common mistakes
Optimizing for views alone
High engagement identifies candidates but does not reveal the hook, proof, audience pain, or commercial relevance that made the content useful.
Using generic analysis prompts
If the prompt does not encode the team's expert evaluation criteria, the model will summarize videos instead of producing actionable creative intelligence.
Skipping human review
The generated brief may be mostly complete, but a marketer must still verify its claims, strategic fit, differentiation, and brand alignment.
Is it for you?
Best for
It is best for e-commerce brands and paid-social teams that need a high volume of evidence-backed creative concepts.
Not ideal for
It is not ideal for teams lacking clear brand context, customer knowledge, or a human reviewer who can judge strategic fit.
From the transcript
“So find find what's popping on organic and then quickly test it as an ad, right?”
“We're doing the research, but the output that we want are new assets, right? New ads. And that generally comes from creating a creative brief,…”
“now we've gone from research to brief in a matter of minutes like literally a matter of minutes”
From the episode
This AI Workflow Replaces 10 Hours of Ad Research