When brands see 27x ROAS, the first instinct is to ask about the audience. What targeting was used. Which lookalike. Which interest cluster unlocked the number. The answer is that none of those things explain it. Targeting is table stakes. Every competent operator has access to the same Meta tools. What drives outlier ROAS is almost always creative.
Meta's own data attributes 49% of purchase decisions on paid social to the quality of the creative itself. That is more than offer, more than landing page, more than audience selection. When a brand is stuck at 2x ROAS and desperately adjusting budgets, bid strategies, and audience exclusions, they are working on the wrong variable.
The creative framework below is what we use with sprint clients. It is not a single piece of advice. It is a five-part operating system that treats paid creative the way a serious brand should: as a repeatable, testable, scalable production process with clear inputs and measurable outputs.
Why most DTC brands have a creative problem they think is a targeting problem
The typical DTC brand running Meta ads produces two to four new creatives per month. They run them against saved and lookalike audiences, watch ROAS for a few days, and conclude that the targeting is wrong when the ads do not convert. The budget shifts. The audiences change. The ROAS stays flat.
The actual problem is creative fatigue and creative poverty. Four creatives per month means the same ads are running for three to four weeks before anything new enters the account. For a brand with any meaningful reach, this means a significant portion of the audience has seen the same creative multiple times. CPM climbs as engagement drops. ROAS falls. The account looks like it has a targeting problem because the symptoms manifest in the targeting data.
Research covering Meta accounts across DTC categories found that brands producing 20 or more net-new creatives per month achieve 65% higher ROAS than those producing fewer than five. The mechanism is simple: more creative volume means more at-bats, higher probability of finding a winner, and a constantly refreshed pool of ads that prevents fatigue from accumulating. The brands with the best ROAS are not the ones with the best audience targeting. They are the ones with the best creative systems.
The 5-part creative framework
Each part is sequential. Missing any one of them will limit the output of the others. The framework works as a system, not as a set of independent tactics.
Audience intelligence: mine before you write
No creative brief should be written before the voice-of-customer (VOC) research is done. VOC research means reading the actual language your customers and near-customers use to describe the problem your product solves, the result they want, and the objections that almost stopped them from buying.
The sources are specific: your own product reviews (especially the five-star and one-star reviews, which contain the clearest language); competitor reviews on Amazon and Trustpilot; Reddit threads in communities where your customer spends time; organic content comments on your own and competitor social accounts; and post-purchase survey responses if you have them running.
What you are looking for is not general sentiment. You are looking for specific language patterns: the phrases people use to describe the before-state and the after-state, the unexpected benefits they discovered, the objections they had before purchasing, and the specific moment or context in which they use the product. This language becomes the raw material for every creative angle in the account. The ads that perform best are almost always using the customer's own words, not marketing copy written in isolation.
The angle hierarchy: match message to awareness
A creative angle is the specific idea a single ad communicates. Not the product. Not the brand. One idea. Most DTC brands run the same two or three angles repeatedly: a product shot with a benefit claim, a lifestyle image with a headline, and occasionally a testimonial. This is not an angle system. It is a lack of one.
Angles should be mapped to where the customer sits in the awareness spectrum. At the lowest awareness level, the customer does not know your product or possibly the problem it solves. The creative needs to lead with the problem or the desired outcome, not the product. At the middle awareness level, the customer knows the problem and is evaluating solutions. The creative needs to differentiate your product from the alternatives they are already considering. At the highest awareness level, the customer has been exposed to your brand and needs a reason to act now. The creative should remove the final objection or create urgency.
A well-built angle bank for a DTC brand has eight to twelve distinct angles across the awareness spectrum, each mapped to a specific customer state and generating a specific response. From this bank, you brief creative. The same product can generate an entirely different ad depending on which angle you select and which awareness level you are targeting. This is why one brand can produce thirty distinct ads without repetition.
Format and platform matching: right creative for the right context
Not every format works at every stage or on every platform. Static images with short copy headlines outperform video on Facebook for retargeting audiences who have already engaged with the brand. Short-form video, particularly content that reads as native to the feed, outperforms static for cold audiences on both Meta and TikTok. Long-form video, 60 to 90 seconds, converts for high-consideration products where the customer needs to understand the mechanism before committing.
UGC-style content, which is creator-led or customer-recorded video designed to look unpolished and native, generates four times the click-through rate of brand-produced creative in most DTC categories. This is not because production quality is irrelevant. It is because highly produced ads signal advertising to the viewer, which triggers the cognitive skip reflex. Native-looking content holds attention because it does not immediately register as an ad.
The format brief should specify: platform, placement (feed versus story versus reel), length, format (static, UGC, voiceover, text-on-screen), and the single action you want the viewer to take. Each of these variables interacts. A 15-second UGC video optimised for Reels placement performs differently from a 15-second brand video optimised for feed. Treating all short-form video as interchangeable is one of the most common creative mistakes in DTC accounts.
Testing velocity: the numbers game most brands lose
Creative testing is a probability game. If your conversion rate on a new creative is 10%, you need to test at least ten to twenty creatives before finding one consistent winner. Most DTC brands run three to five, conclude that creative testing is not working, and go back to their existing ads. They have not tested enough to find the winner. They have given up before the numbers could work in their favour.
The minimum viable testing cadence for a brand spending between £5,000 and £20,000 per month on Meta is eight to twelve net-new creatives per month across three to five angle categories. This does not require a large production budget. A creator network of three to five people receiving product and a brief can produce eight to twelve assets per month at a total cost of £800 to £2,000. The brief system described in parts one and two reduces the creative direction burden to near zero once it is set up.
Testing velocity also requires a clean account structure. Each new creative should go into a dedicated ad set with a fixed budget and a fixed audience so you can isolate creative performance from audience variables. Running new creative in existing ad sets with mixed historical data produces uninterpretable results. The test reads as the audience, not the creative.
The iteration loop: winner signals and kill rules
Most brands do not have explicit criteria for what makes a creative a winner or what triggers a kill decision. As a result, ads run until the budget runs out or the ROAS drops to an unacceptable level. By the time the kill decision is made, the ad has spent weeks burning budget at declining efficiency.
The winner signal is a combination of three data points within the first five to seven days: CTR above your account baseline (typically 1.5% to 2.5% for cold audiences), a cost-per-click at or below your target, and at least ten purchase events attributable to the ad. An ad that clears all three thresholds gets scaled: moved into a higher-budget campaign with broader audiences and additional creative variations built from the same angle.
The kill rule is simpler: if an ad has spent the equivalent of one to two target CPAs without a purchase, it is dead. Not underperforming. Dead. Do not wait for more data. Do not adjust the creative or the audience. Kill it and move the budget to the next test. The speed of the kill decision is as important as the speed of the scale decision. Slow kill decisions are the primary way DTC brands lose money in creative testing.
What 27x actually looked like
The 27x ROAS result came from a wellness consumable with a high repeat purchase rate and a low product cost relative to the retail price. The margin structure was favourable before the creative work started. But the brand had been running Meta ads for eight months at between 2x and 3x ROAS with the same five creatives cycling. The ads were polished. The messaging was generic. Nothing broke through.
The first intervention was VOC research. Three hundred product reviews and sixty Reddit comments later, a specific language pattern emerged: customers were not primarily buying for the stated benefit on the packaging. They were buying to solve a secondary problem that the brand had never acknowledged in its advertising. A creator brief was written around that problem, leading with the exact language customers used to describe it in their own words.
Two UGC videos were shot from that brief. Both were under 30 seconds. Neither featured any product packaging. The hook in both was the customer's problem stated in the customer's language. One of the two videos generated a 6.8% CTR against a category norm of under 2%. It was scaled within 72 hours. Over the following four weeks, with the full five-part framework running, the account ROAS reached 27x.
27x is not a repeatable daily run rate. It was a peak. But the account maintained above 8x for the following six weeks while the creative system continued finding the next set of winners. The underlying reason is not complicated: the framework ensured a constant supply of new angle-driven creative, tested systematically, killed quickly when they failed, and scaled aggressively when they won.
The build order if you are starting from scratch
If your current creative system is ad hoc and you are building the framework from zero, the practical sequence is: VOC research first (two to three days if you are thorough), angle bank second (document eight to twelve angles before briefing anything), format decisions third (lock your platform and placement strategy before production begins), then brief and produce your first eight to twelve creatives, launch with clean test ad sets, and set your winner and kill criteria before the tests start running.
The briefing and production cycle runs continuously after that. Every two weeks, a new batch of eight to twelve creatives enters the system. Every week, you are reviewing performance data and making scale or kill decisions. The framework becomes self-funding quickly: winners generate the ROAS that funds the next round of testing.
The time to first signal on whether the framework is working is four to six weeks. At that point you will have run two batches of creative, made a number of kill decisions, and probably found one or two genuine angle-level winners. By week twelve, the compound effect of systematic testing will be visible in the account data. ROAS will be trending up. CPM will be stable or falling. Creative fatigue will be a non-issue because you are constantly refreshing the pool.
Find out where your creative is leaving ROAS on the table
The Brand Growth Audit covers your full paid media setup: creative angle analysis, account structure, testing velocity, and the specific levers that will move your ROAS in the next 30 to 90 days. Three days, Loom walkthrough, prioritised PDF.
If you want a faster read on where you sit before committing to an audit, the free scorecard includes a creative and paid media section that will give you a benchmark comparison in under three minutes.
Frequently asked questions
What is a good ROAS for DTC brands on Meta?
A good Meta ROAS for a DTC brand depends on your contribution margin. At a 50% contribution margin, you need at least 2x ROAS to break even on ad spend once COGS and other costs are accounted for. Most DTC brands operating profitably on Meta run between 2.5x and 4x blended ROAS. Anything above 5x consistently is excellent. 27x ROAS is an outlier result achieved through a combination of a high-margin product, a very low CPM environment, and a creative system that generated outsized click-through and conversion rates simultaneously.
How many creatives should a DTC brand test per month?
Research shows that DTC brands testing 20 or more new creatives per month achieve 65% higher ROAS than brands testing fewer than five. Most brands producing under ten creatives per month are running on creative fatigue and do not realise it. A realistic minimum for a brand spending between £5,000 and £20,000 per month on Meta is eight to twelve net-new creatives per month.
What is creative fatigue and how do I know when it is happening?
Creative fatigue occurs when an audience has seen a particular ad enough times that engagement and conversion rates begin to decline. On Meta, the clearest signal is a rising CPM paired with a falling CTR on ads that previously performed well. Frequency above three to four within a seven-day window for a given ad set is typically where fatigue accelerates. The fix is not audience expansion, it is fresh creative.
Should DTC brands use UGC or brand-produced ads?
Both have a role, but UGC consistently outperforms brand-produced creative on CTR, often by two to four times, because it reads as native content rather than advertising. For DTC brands on Meta and TikTok, UGC-style content should represent the majority of your testing volume. The right split for most DTC brands is roughly 70% UGC and creator-led content, and 30% brand-produced.
How long does it take to see results from a creative testing framework?
The first clear signals typically emerge within three to four weeks, assuming sufficient ad spend to generate statistically meaningful data. At £5,000 to £10,000 per month, you should have enough signal within four weeks to identify one or two genuine winners from a batch of eight to twelve creatives. By week eight, a brand with a functioning creative framework should see measurable improvement in blended ROAS versus the pre-framework baseline.
What should a DTC brand creative brief include?
A strong DTC creative brief should include: the customer segment being targeted and their specific pain point or desire; the single message or angle the creative must communicate; the hook format (problem-led, result-led, curiosity-led, or testimonial); the call to action and where the viewer lands; reference creative that performs in the account or the category; and any non-negotiables on brand presentation. The brief should be short enough to read in under two minutes.
About the author
Caner Veli founded and exited Liquiproof, scaling from zero to 3,000+ retailers globally in under 6 years. He now runs Purposeful Profits, a focused growth consultancy for founder-led DTC and CPG brands. 12 named sprint clients. 518% average growth. 27x highest ROAS. Read more about Caner →