Meta Ads in 2027 | The Creative Playbook That Actually Works

Summary

The presenter argues that Meta ad performance increasingly depends on the quality and diversity of creative—not complex ad-account structures. One client’s ad reportedly ran for a year and continued to receive spend. With Meta’s AI influencing distribution, brands need a range of relevant formats, messages, angles, and personalities. The goal is not simply to upload more ads, but to make each creative more purposeful.

Creative strategy playbook

• Start with customer research. Identify customers’ pain points, desires, objections, and the language they use. Review your own and competitors’ reviews, comments on ads, and discussions on Reddit. Treat negative comments and reasons for not buying as useful research.
• Use AI tools to analyze research. The presenter uses a Slack-based tool referred to as Runna by Motion to surface recurring objections. He also recommends giving tools such as Claude or ChatGPT customer reviews, ad comments, and other brand context before asking for research or script drafts.
• Build each concept from three elements:
  1. Angle or message: What is the ad saying?
  2. Customer persona: Who is it speaking to?
  3. Format: How will the message be delivered?
• Match creative to awareness level. An unfamiliar prospect needs different messaging from someone close to buying. Use varied creative to move people from initial awareness toward purchase and loyalty.
• Use social feeds to spot formats and language. The presenter describes creating separate social accounts for buyer personas and browsing them to see the content, hooks, and formats those audiences encounter.

Hooks, scripts, and visuals

A hook can combine a visual event, spoken line, and on-screen text. Start with a product-relevant problem and create curiosity that the rest of the ad resolves. “Mirror hooks” reflect the audience’s identity, questions, or frustrations.

The presenter says hooks still matter, but for longer ads, retention may be more important than opening engagement alone. Judge a hook in the context of the ad’s overall conversion performance.

A recommended problem–solution script structure is:

1. Hook
2. Problem
3. Intensify the problem
4. Solution and benefit
5. Proof or social proof
6. Call to action

Other useful structures include founder stories, educational scripts, and “villain vs. hero.”

Make the product part of an engaging, natural narrative rather than leading with a hard sell. The presenter suggests that delaying the product mention can make an ad feel less sales-like.

Match visuals to both the format and the script. For UGC reviews, the speaker favors showing the person speaking while using or demonstrating the product, rather than frequently cutting to unrelated footage. For VSLs, illustrate each key line with a relevant shot.

For example, in a self-tanner ad, a line about avoiding stains on white bedding could be illustrated by a hand leaving a stain on a white sheet. In an example for The Black Stuff, shots of soap production and the founder support the claims being made.

Formats and channel tactics

Formats the presenter says are performing well include UGC, founder ads, mini-VSLs, and podcast-style ads.

• Affiliate and co-branded ads: Meta’s co-advertising or whitelisting approach can add social proof and make an ad feel more native. Many clients reportedly run the same creative from both the brand account and as an affiliate ad.
• Static ads: Use them to capture lower-funnel demand with concise comparisons, testimonials, problem–solution messages, or brand reminders. The presenter describes them as inexpensive to produce and recommends having a deliberate system for creating them.
• Creative diversity: Mix formats and messages rather than relying on one polished style. Examples range from simple, low-budget UGC to higher-production podcast ads.

AI and production

Use AI to speed up research, drafting, image creation, voiceovers, or video—but provide strong brand and customer context. The presenter warns that generic AI output can make a brand blend in with the volume of mediocre AI-generated ads.

He recommends using AI to increase productivity and, ideally, quality—not to replace strategy. Suggested lower-risk creative applications include cartoon-style static ads and product imagery rather than synthetic depictions of people.

The presenter cautions that laws and regulations are changing and specifically mentions new New York laws expected in 2027 affecting AI in advertising. He presents this as a reason to check applicable rules before using generated content.

Production options include working with an agency or filming in-house. The speaker says modern phones are adequate for founder videos, unboxings, and UGC-style ads, and encourages brand teams to make ads themselves to learn from the process.

For casting, he recommends outreach on Instagram and TikTok and maintaining a database of creators who have already worked with the brand.

Testing and measurement

Ensure each ad gets enough opportunity to be evaluated. The presenter warns that putting everything into CBOs and forcing ads to compete can allow only a few creatives to consume most of the budget. Minimum spend levels can help gather evidence about weaker ads as well as potential winners.

Avoid combining or “mixing” creative elements in a way that obscures what drove performance. Analyze results so the team can learn which component worked.

Track hook rate, click-through rate (CTR), hold or retention rate, engagement rate, and conversion rate. Assess downstream conversion—not just engagement or retention—to determine whether a format is profitable. For example, podcast ads may have strong retention but lower engagement; conversion data is needed to establish whether that retention produces customers.

Examples and figures mentioned

• A client’s ad reportedly continued running and receiving spend for 12 months.
• The speaker suggests testing four or five ads per week, with different messages, customer profiles, and formats, as one way to build a balanced creative mix.
• The agency offer is aimed at brands spending over $70,000 per month; no currency is specified in the subtitles.
• Examples include The Black Stuff (soap products and recurring price objections) and Groom’s (a customer who switched from AG1 and disliked the taste of green drinks).

Presenter and sources

• Presenter: Fraser Cottrell
• Brands and examples: The Black Stuff, Groom’s, AG1
• Platforms and tools referenced: Meta, Reddit, Runna by Motion, Motion, Claude, ChatGPT, Slack, Instagram, and TikTok