Most dropshipping creative advice lists angles and stops there. The list is the easy part. The part that decides whether you make money is how much you spend before you trust the number, and almost nobody puts a figure on it.
So here is the figure first: about three times your target CPA per angle. At a modeled $35 CPA for apparel, that is $105 to test one angle to the point where the result means something. Seven angles is $735.
If that sounds like a lot, the alternative is worse. Testing at $30 an angle produces numbers that reverse themselves on a single conversion, and decisions made on those numbers are coin flips with extra steps.
The economics underneath
| Segment | CTR | CPA | ROAS | Lifespan |
|---|---|---|---|---|
| Fast fashion / basics | 1.45% | $26 | 3.9x | 6 days |
| Activewear | 1.30% | $32 | 3.7x | 8 days |
| Footwear | 1.15% | $38 | 3.4x | 9 days |
| Denim | 1.05% | $42 | 3.3x | 10 days |
| Premium / outerwear | 0.80% | $58 | 2.9x | 11 days |
Cheaper products convert cheaper and fatigue faster. That relationship holds across the whole set and it means your test budget and your refresh cadence are both functions of price point, not of how good your creative is.
One number outside the table matters more than anything in it: 30-day retargeting runs a modeled 6.4x ROAS against 2.6x for cold prospecting. Before testing a seventh prospecting angle, check that retargeting is running at all. The full set is on our fashion and apparel benchmarks.
Seven angles, with test cost and kill threshold
1. Problem-first
Open on the irritation the product removes, before the product appears. Reliable because it selects for intent rather than aesthetics.
Test cost: $105. Kill at: 3x CPA, zero purchases.
2. Unboxing on a phone
Someone opening the package, filmed badly on purpose. Cheap to produce and it survives the fatigue curve better than studio product footage.
Test cost: $105. Note: the cheapest angle to produce, so test it early.
3. Before and after
Works in a narrow set of categories and falls flat everywhere else. If your product does not produce a visible change, skip this one rather than forcing it.
Test cost: $105, but only in categories where the change is genuinely visible.
4. Comparison against the obvious alternative
Yours next to the thing the customer would otherwise buy. Sharp, and it front-loads the objection instead of waiting for it.
Test cost: $105. Rotate: which alternative, every fortnight.
5. Use-case stack
Five contexts in eight seconds. Sells versatility, which is what turns a maybe into a purchase for mid-price items.
Test cost: $105. Best at: $40 and above, where the buyer needs more than one reason.
6. Social proof screenshot
A real review, screenshotted, at hero size. Works well and carries an obvious constraint: you need real reviews. Do not manufacture them.
Test cost: $105. Prerequisite: genuine review volume.
7. Price-anchored offer
The number, large, with the reason it is that number. Fatigues fastest of the seven, in line with deal creative across every vertical we model.
Test cost: $105. Kill at: 2x CPA rather than 3x, since decay is quick enough that a slow test is a wasted one.
Reading the result without fooling yourself
Three outcomes, and only one of them is a decision.
- Zero purchases at 3x CPA. Dead. Kill it and move on. - One purchase above target cost at 3x CPA. Inconclusive, not dead. Fund one more increment before deciding. - At or below target CPA. It works. Now the real job starts, because an 8-day lifespan means you are rebuilding it in a fortnight.
The expensive mistake is killing at 1x CPA. At that spend a working angle and a dead one look identical, and the accounts that plateau are usually the ones that discarded winners they never funded properly.
What happens after an angle wins
An angle that survives testing is not an asset, it is a template. Fast fashion burns a creative in 6 days and activewear in 8, so a winning angle needs a fresh execution roughly weekly: same structure, different hook, different opening frame.
Rotate the hook rather than the product. That is the pattern across accounts that hold ROAS through a quarter, and it is the reason a working angle can carry months of spend while the individual ads inside it last a week.
The volume this implies is the actual constraint for most dropshipping operations, and it is what our dropshipping workflow is built around. Plans are on pricing.