diff · your funnel · before / after
buy more traffic.
redesign the whole site on a hunch.
spend another dollar on media.
more revenue from the visitors you already paid for.
FIXED.
CRO is what turns existing traffic into more revenue without spending another dollar on media. Most brands leave the biggest wins in their funnel (PDP, cart, checkout, upsell) on the table. We don’t. We find the leaks, and we fix them. Month after month.
Somewhere in your checkout, right now, somebody just gave up.
EXHIBIT · WHERE THE MONEY ESCAPES
While you sleep, it leaks.
Every visitor in your funnel has already been paid for, by media, by SEO, by every other line of the budget. The leak isn’t a metaphor on this page. It’s the gap between the sessions you bought and the orders you kept.
- LEAK 01PDPThey look. They leave.
- LEAK 02CARTThey add. They abandon.
- LEAK 03CHECKOUTThey start. They stall.
- LEAK 04UPSELLThey buy once. They were never asked twice.
The biggest wins are sitting exactly here, on the table. Our job is to take them off it.
LOOP 01 / 05 · AUDIT
FIRST, WE WATCH.
Quantitatively and qualitatively, at the same time. The analytics tell us where the funnel bleeds. The replays, the heatmaps and the user tests tell us why, in your customer’s own behaviour, not in anyone’s opinion.
- QUAL · SESSION REPLAY
She scrolled past the size chart four times, then rage-clicked the one that never opens.
- QUAL · HEATMAPS
Everyone taps the product photo. The product photo does nothing.
- QUAL · USER TESTING
“I just wanted to know if it ships free. I gave up looking.”
LOOP 02 / 05 · PRIORITISE
THIRTEEN TESTS. RANKED. NOT VOTED.
Every candidate fix is scored by expected impact and effort, never by the loudest idea in the room.
- 01PDP3 TESTSprovenance · social proof · delivery
- 02CHECKOUT4 TESTSApple Pay · guest · shipping logic · PO
- 03HOME3 TESTShero · quiz · category logic
- 04CART3 TESTScross-sell · free-shipping bar · trust badges
the loudest idea in the roomunranked. unscored. unshipped.
LOOP 03 / 05 · TEST · STRUCTURED A/B, PROPER STATISTICAL RIGOUR
two truths enter. one ships.
HYPOTHESIS · 2026-041Leading with provenance (“Hand-forged, Michigan, 1974”) will beat leading with price.
- CVR
- 2.14%
- AOV
- $168
- SESSIONS
- 14,820
- ORDERS
- 317
- CVR
- 2.91%
- AOV
- $184
- SESSIONS
- 14,960
- ORDERS
- 435
LOOP 04 / 05 · SHIP · WINNERS SHIPPED, LOSSES DOCUMENTED
the register · last eight verdicts
Eight tests. Five shipped. Two killed. One sent back for rerun. The log, not the loudest person, decides what stays.
LOOP 05 / 05 · REPORT · MONTHLY, ON WHAT MOVED AND WHY
Then back to 01. The reported month becomes the next audit; the next leak is already on the list. That’s the whole trick: there is no finish line, only a higher baseline.
WHAT YOU DON’T DO
- You don’t argue about which button to change.
- You don’t read statistical significance reports.
- You don’t chase every CRO tip on LinkedIn.
The arguing, the statistics, the discipline of ignoring the feed. That’s the work. Ours.
WHAT YOU GET
- SLIP AA prioritised test roadmap.ranked by impact and effort, never by the loudest idea
- SLIP BMonthly test results.what moved, what didn’t, and why, read properly
- SLIP CA running document of what we’ve learned about your customer.every verdict teaches; the document compounds like the wins do
Every opinion about your site is one test away from being wrong, or right.
RUN IT.
We’ll build the backlog, run the tests, retire the opinions, and keep the winners stacking. Your funnel has leaks tonight. Let’s go find the first one.
the wingman project · run · §III.09 · cro · loop returns to 01