'$6 off' vs '15% off' on an upsell: 13.59% vs 10.3% take rate
Commentary byLiftArchiveOur analysis, not theirsFraming a post-purchase upsell as "$6 off" instead of "15% off" raised the take rate from 10.3% to 13.59% and added $0.60 per visitor.
Ezra Firestone ran the test on the BOOM! by Cindy Joseph store and published it on the Zipify blog.
The dollar version was also a bit cheaper, so framing is not the only difference.
- Take rate: 13.59% for $6 off vs 10.3% for 15% off.
- Revenue per visitor: +$0.60 for the dollar version, even though each accepted order brought in less.
- A third test in the same post on another product, 20% off vs $9 off, gave $0.73 vs $0.51 per visitor for the dollar version, at nearly the same price.
- Visitor counts and confidence levels were not shared.
The change
Discount shown as $6 off ($22) instead of 15% off ($23.80).
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
After buying a Boomstick Color, buyers were offered another one at 15% off, paying $23.80 instead of $28.
Variant
The same offer was shown as $6 off, for a price of $22. Everything else on the page was identical.
Why it likely won
Our reading of the result, not the practitioner's claim.
- The $6 offer was $1.80 cheaper, which on its own could lift the take rate.
- A dollar figure tells buyers exactly what they save, with no mental math.
- The third test, where prices were only 20 cents apart, points the same way, which suggests the framing itself matters.
Caveats: The dollar version had a lower price, so price and framing are mixed together in this test. Vendor-founder test with no sample size or confidence shared.
Who ran it
Ezra Firestone ran this test for BOOM! by Cindy Joseph and published the results (Zipify blog). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Ezra Firestone →
How to run this test on your store
- Hypothesis
- If we show the upsell discount as a dollar amount instead of a percentage, at the same final price, then revenue per visit will rise because a concrete saving is easier to judge.
- Primary metric
- Revenue per upsell page visitor
- Guardrails
- Upsell take rate
- Margin per upsell order
- Average order value
- Who should test it
- Stores running post-purchase upsells on products priced roughly $20 to $50. Keep the final price equal in both versions, or you won't know whether framing or price drove the result.
Questions
Was this a fair test of dollar vs percent framing?
Not fully. The dollar offer was also $1.80 cheaper.
A separate test in the same post, with prices 20 cents apart, also favored the dollar version.
Does a higher take rate always mean more revenue?
No, but here it did. Each order earned slightly less, yet enough extra buyers said yes to add $0.60 per visitor.
Similar tests
Vasta
Visually