Buy-three bundle message by add-to-cart: -7.9% AOV, inconclusive
Test run byClean CommitCRO agency
Commentary byLiftArchiveOur analysis, not theirsShowing a "buy three, get money off" message beside the add-to-cart button lowered average order value 7.9%, the metric it was meant to raise.
Clean Commit ran this test for an anonymized Shopify store.
With only 1,153 visitors over 7 days, we treat the result as inconclusive.
- Average order value fell 7.9%, the opposite of the goal. The control had an 81.9% probability of the higher order value.
- Revenue per visitor fell 11.4%. This is negative, but the sample is too small to trust.
- Add-to-cart rate rose 7.0% and bounce rate fell 11.3%, so the message drew interest.
- Conversion rate moved -1.9%, which is no clear difference.
The change
The variant adds a text-only multi-buy discount message by add-to-cart.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The product page had no multi-buy message near the add-to-cart button.
Why it likely came out this way
Our reading of the result, not the practitioner's claim.
- A message alone asks shoppers to work out the deal and add three items themselves, which few may bother to do.
- The discount may have pulled in shoppers who added items but then checked out with fewer, or smaller, orders.
- With about a thousand visitors, mostly on mobile, random swings can easily explain these moves.
Caveats: Very small sample: 1,153 visitors over 7 days, mostly on mobile. Read the result as directional only.
Who ran it
Clean Commit ran this test for a Shopify store (anonymized) and published the results (Clean Commit A/B test library). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Clean Commit →
How to run this test on your store
- Hypothesis
- If we show the multi-buy discount next to the add-to-cart button, then average order value will rise because shoppers see the offer at the moment they decide to buy.
- Primary metric
- Average order value
- Guardrails
- Conversion rate
- Revenue per visitor
- Who should test it
- Stores selling consumables or items people buy in multiples, with enough traffic to run for several weeks. If you can only get a thousand visitors a week, test something bigger first.
Questions
Does a multi-buy message near add to cart raise order value?
Not in this small test: order value fell 7.9%.
The sample of 1,153 visitors was too small to say for sure either way.
Should a bundle offer be text or a selectable option?
This test used text only. A selectable bundle that adds all items at once removes effort, so it is worth testing as an alternative.
How long should a bundle offer test run?
Long enough to reach a stable result on order value, which varies a lot.
A week on about a thousand visitors was not enough here.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon