Case Study
Super Area Rugs: 216% revenue increase in 37 days
Last updated: [Updated Date]
216.29% revenue increase in 37 days, no extra ad spend
The challenge
Super Area Rugs is an ecommerce DTC store in the home-and-decor category. The brief was specific: revenue increase from existing traffic. No new paid spend. No new audience acquisition. Pure CRO.
Stores at this revenue band typically have inherited a CRO programme that has been running on autopilot. Quarterly tests, 95% significance defaults, a hypothesis backlog generated by whatever the AI tool surfaced. The lift band sits in the 4-7% range. Comfortable. Not interesting.
The 216% figure required something structurally different.
What I did
37 days. Multiple tests compounded inside the window. The pattern:
Each individual test ranked by The Evidence Stack before it shipped. Hypotheses pulled from the existing backlog were re-scored. Anything below 6/20 on the stack was killed before it consumed traffic. The remaining tests had a much higher prior probability of producing a real winner.
Winner-calls gated at 99% confidence (The 99 Rule). Industry-default 95% means a 1-in-20 false-positive rate. Compounding 8-12 tests across 37 days at 95% means 30-40% of the "winners" are statistical noise. At 99%, the false-positive rate is 1-in-100 — winners actually win.
Failure-as-information. The losing tests fed the next round. Every losing variant told us something specific about which hypothesis bucket to deprioritise next.
The outcome
216.29% revenue increase from existing traffic in 37 days. No additional ad spend. No new audience. Same store, same products, same pricing.
What this proves
The compounding-tests-at-99% pattern produces results that look impossible from outside. From inside the methodology, they're the natural output of removing the false-positive noise that drags down most CRO programmes.
The 4-to-34 Gap (per The 4-to-34 Gap) is the band between DIY-AI-tool lifts (4-7%) and operator-led-AI lifts (28-34%). Super Area Rugs sits at the upper end of that band, compounded across multiple test cycles inside a single engagement window.
Want a similar engagement?
The Sprint engagement at £2,500 is the right shape for a 30-day compounding test programme. Book a free 15-minute CRO audit for an honest read on whether your store fits.
Want this kind of result on your site?
AI Headline Lab £500 (5 working days) is the productised version of the work above. AI-generated headline variants, ranked against your buyer’s actual language, ready to test against live traffic.
Different lever? Speed Sprint £1,500 (2 weeks) if your site fails Core Web Vitals, or Pricing Experimentation Audit £2,500 if your promo windows leave revenue on the table. Or book a free 30-minute audit to scope the right entry product for your site.
