New video ad in a testing campaign: $8.43 cost per purchase, ROAS 4.50
Commentary byROASArchiveOur analysis, not theirs@EcomRedneck added a new video ad to their main Meta testing campaign and, the next day, shared a screenshot showing a cost per purchase of $8.43 and ROAS of 4.50. It was an early, partial-day read with no comparison ad, so these are one ad's own numbers, not a measured lift.
- Cost per purchase: $8.43.
- ROAS: 4.50, which works out to about $4.50 in tracked revenue for each $1 of spend.
- Purchases: only a few, from part of one day (the screenshot is a single ad row). Spend and the full day's numbers are not shared, so we can't tell how stable these numbers are.
The change
Before
@EcomRedneck adds a few new ads to their main testing campaign every day. The post doesn't share how earlier ads or formats performed.
After
A new video ad added to that testing campaign. The day after launch it was taking spend and, in the author's words, doing very well.
Our reading: why it likely worked
Our interpretation, not the advertiser's claim.
- Video can show the product in use and answer objections in seconds, which a static image often can't.
- Adding a few new ads to a testing campaign every day keeps fresh creative coming, so a possible winner shows up quickly.
- One ad on part of one day is a tiny sample, and the post doesn't compare it with earlier ads, so we can't say video beat anything.
Caveats: One ad in one ecommerce account, read on part of a single day with only a few purchases and no control. Spend, the final daily numbers and the attribution window are not shared, and income results like this vary and are not typical.
How to run this in your account
- Hypothesis
- If we run new video ads in their own testing campaign, then we will find winners at a lower cost per purchase because video shows the product better and each ad gets its own budget.
- Primary metric
- Cost per purchase
- Guardrails
- ROAS
- Hook rate (3-second views)
- Spend per ad
- Who should try it
- Ecommerce brands running mostly static ads, or with no structured way to test new creative. It may not pay off if you can't produce videos regularly.
Questions
Do video ads perform better than images on Facebook?
This new video ad reported $8.43 per purchase and 4.50 ROAS on an early, partial-day read, but there was no image ad to compare against.
Run both side by side to know for your account.
Should I test new ads in a separate campaign?
A testing campaign gives new ads budget without competing with proven ones.
Here a new video ad added that way showed a 4.50 ROAS on its first partial day.