Mums Candy: what changed, and what it paid.
21 of 38 tests won, and $1.34M in incremental revenue
20 minutes · your numbers · no deck
Mums Candy · Confectionery · DTC ecommerce · A/B testing
$1,338,625
incremental CRO revenue
- 9.8% conversion rate, up 162.7%
- −38% cost per acquisition, year on year
- 21/38 A/B tests won
What was broken, and what changed
The challenge
- Value proposition unclear above the fold: visitors could not tell quickly what made the product worth buying
- Paid traffic landing on pages that never repeated the promise the ad made
- Impulse framing weak for a category that is bought on impulse
- Trust signals buried below the decision point
- Product detail pages carrying friction at every step
What we did
- Rebuilt the value proposition so the offer is legible in the first screen
- Moved trust signals up to sit at the moment of decision, not after it
- Stripped friction out of the product detail page step by step
- Reframed the offer around urgency the category actually has
- Put a continuous A/B testing programme underneath all of it
From research to a hypothesis
What the research found
The product is an impulse buy and the page was not built like one: the order was the last decision it asked for, after several smaller ones.
Paid traffic made it worse: the ads promised one thing, the landing experience opened with another, and the visitor had to do the work of connecting them. Most did not.
The hypothesis it tested
If the offer is legible in the first screen and every trust signal sits at the moment of decision rather than after it, an impulse category converts on the traffic it already has.
The result, on the record
Variants run against a control on live traffic; the count is wins over tests run.
Mums Candy · results
- incremental CRO revenue
- $1,338,625
- Before · conversion rate before the work
- 3.7%
- conversion rate, up 162.7%
- 9.8%
- cost per acquisition, year on year
- −38%
- A/B tests won
- 21/38
- Timeframe
- Ongoing programme
- Tests run
- 38
- Beat control
- 21
- Lost or inconclusive
- 17
Basis: controlled tests. What each kind of evidence can prove →
“Solvexmedia unlocked revenue we were leaving on the table — measurable growth and real ROI.”
If your store has the same problem
Twenty minutes on your numbers: whether the same leak is there, what we would fix first, and whether the work is worth buying at your traffic.
20 minutes · your numbers · no deck
The full write-up
Mums Candy had the traffic. Paid was working in the sense that people arrived; it was not working in the sense that they bought.
What the research found
The product is an impulse buy. The page was not built like one. The value proposition sat below the fold, the trust signals sat below that, and the product detail page asked for several small decisions before it asked for the order.
Paid traffic made it worse. Ads promised one thing, the landing experience opened with another, and the visitor had to do the work of connecting them. Most did not.
What changed
The rebuild put the offer in the first screen and moved every trust signal up to the decision point rather than after it. The product detail page lost a step at a time, each removal tested, not assumed.
Then the programme: thirty-eight tests, run properly, sized against the traffic before anything was built.
What happened
21 of 38 tests won. Which is the number worth dwelling on: seventeen did not. Every one of those was reported, and each eliminated a branch of the backlog that would otherwise still be open.
Conversion rate moved from 3.7% to 9.8%, a 162.7% increase. Cost per acquisition fell 38% year on year, on the same traffic sources. Cumulative incremental revenue attributable to the programme: $1,338,625.
No increase in ad spend. The traffic was always there.
More like this one
What happens after you book
- Pick a time Twenty minutes, camera optional.
- We look at your store live Your pages and your numbers, not a deck.
- You leave with the plan The three leaks we would fix first, whether or not you hire us.
Want the Mums Candy version of this on your store?
Twenty minutes on your numbers. We tell you where you are leaking, what we would fix first, and whether the work is worth buying at your traffic.
Got it. Now pick a time.
Open the calendar →Twenty minutes on your store and your numbers. If we cannot help we will say so.
- 437+ stores built
- 2,000+ tests shipped
- No call needed to start