The $100/Day Creative Test Framework for DTC Brands

Three days at $20/day per concept give you engagement signals. CTR. CPC. Add-to-cart rate. These are not statistical signals. They are early-stage attention signals.

Concepts that get no attention at $20/day will not magically get attention at $200/day. The slope of the curve does not change. A 0.4% CTR concept is dead at every spend level. Killing it on day 3 saves you $1,400 over the next 30 days. That is what $100/day testing earns you. Speed. Not certainty.

Three rounds, each with a specific job.

• Week 1, Concept Tournament. Ship 5 different concepts. Run them in parallel. Kill the bottom 2 on day 3.

• Week 2, Variation Round. Take your top 3 concepts. Make 2 variations of each (different hook, different opening, different CTA). Run all 9 in parallel. Kill the bottom 6.

• Week 3, Scale Test. Take your top 2-3 variations from Week 2. Push budget to $40/day per variation. See if the engagement signals hold when the audience pool widens.

Three weeks, 14 individual creatives shipped, somewhere between zero and three real winners identified. Most rounds will leave you with one. That is a successful round.

The job in Week 1 is concept-level differentiation. You are testing whether each concept has any pull on a cold audience.

Ship 5 concepts. Structurally different concepts, not the same ad with different colors or hook text.

A reasonable Week 1 lineup for a DTC product:

Budget allocation: $20/day per concept, run for 3 days. Total Week 1 spend: $300.

End-of-day-3 review. Apply the kill criteria:

• CTR below 0.8% \= kill. The concept is not earning the click.

• CPC above $2 \= kill (for sub-$80 AOV products). Attention is too expensive.

• ATC rate below 1% \= kill (if the concept got clicks but they bounced).

These numbers shift by category. A $300 AOV product tolerates a $4 CPC and still works. A $35 AOV product does not. The framework matters more than the exact thresholds. Pick thresholds appropriate to your AOV before the test starts. Write them in the template before the test starts so you do not move the goalposts mid-round.

Three concepts survive Week 1. Two die. Move to Week 2.

You have three concepts that earned attention. Week 2 sharpens them.

For each surviving concept, ship 2 variations. Pick one variable per variation. Common variables:

• Hook variation. Same concept, first three seconds rewritten. The hook drives roughly 70% of CTR variance in our experience.

• Opening visual variation. Same hook, different opening shot or product framing.

• CTA variation. Same concept, different call-to-action language ("shop now" vs. "see why" vs. "try one").

Now you have 9 ads running. At $100/day total, that is about $11/day per variation. Run for 3 days.

End-of-day-3 review. Apply slightly tighter criteria (since you are now comparing variations of concepts you already know have pull):

• CTR below 1.0% \= kill.

• CPC above $1.50 \= kill (sub-$80 AOV).

• ATC rate below 1.5% \= kill.

Kill 6, keep 3. If two of your kept variations are from the same parent concept, that is fine. Move both into Week 3.

This is the only week of the three where you are doing what most operators think of as "testing." You take the 2-3 variations that survived Week 2 and push budget to $40/day per variation. Total weekly spend: $700-840.

Why this week matters: a concept that performs well at $11/day on a small, niche audience often degrades when the algorithm has to push into broader, more expensive impression space. Week 3 is where you find out which winning variations survive scale.

Run for 5-7 days. Apply scale-test criteria:

• CTR holds within 20% of Week 2 performance \= the concept scales.

• CPC rises less than 30% \= the audience widens cleanly.

• ATC rate holds \= the post-click value of the concept survives.

A concept that fails Week 3 is not a bad concept. It is a niche concept. Keep it in your library for retargeting or use it again with a tighter audience setup. Do not put your main spend behind it. Move spend behind the variations that hold up.

Three weeks. Roughly $1,500-1,800 in test spend. One to three concepts validated for scaling. The next month of your account runs on what you learned in those three weeks.

This framework will not give you statistical confidence. It is not a winner-picker. It is a loser-killer. The difference matters.

At $100/day you are working with engagement signals (CTR, CPC, ATC), not conversion signals. Conversion data is too sparse to be reliable at this scale. If you only get 4 conversions on a concept in three days, you have no idea whether that 4 is a 1.2% conversion rate or a 4.8% conversion rate hiding under sample noise.

What you do get: a fast read on which concepts are dead on arrival. Concepts with bad engagement signals will not bloom into winners at scale. The algorithm does not save a bad ad. Killing those concepts early lets you redirect spend toward the concepts that earned the click.

Two rules to keep yourself honest.

The 3-week framework is the small-spend version of what we wrote about in [the Creative-First CAC Curve](/contact). At any spend tier, the dominant variable in paid performance is creative volume and variety. At $100/day, you do not have the budget for variation-at-scale testing. You have the budget for fast killing.

Fast killing is the right job for your spend tier. By the time you graduate to $200, $400, $1000/day budgets, you will already have a creative pipeline that ships and kills concepts on a regular cadence. That habit is more valuable than any single winning ad you find.

The tracking template is the spreadsheet version of this framework. Concepts, variations, CTR, CPC, ATC, kill thresholds, scale-test results, all in one tab. Type in your numbers as the test runs. The template tells you what to kill and what to keep.

[Get the creative test tracking template](/contact). Free. Email-only signup.

For the bigger picture on how creative drives CAC at every spend tier, read [the Creative-First CAC Curve](/contact). That post is the framework this one operationalizes at the small-DTC budget tier.