MarketrixMarketrix1,304 of 2,400 simulated founders chose the outcome-led hero. The lift is concentrated in first-time, non-US and support-seeking personas.
Both variants showed the same page below the fold; only the hero differed. Personas that decide on eligibility and speed moved to B. Personas that decide on cost were indifferent, and trust-first personas moved slightly toward A. Treat the lift as real for the audiences it came from, not as a blanket win.
Scored across eligibility, cost, safety, spend oversight, company fit, and uncertainty.
| Dimension | CURRENT HERO | OUTCOME-LED HERO |
|---|---|---|
| Clarity of who it's for | Feature-led; startup fit implied | Outcome-led; speed and eligibility stated |
| Trust signal | Measured tone reassured risk-averse founders | 'Afternoon' read as rushed to trust-first founders |
| Cost transparency | Same table below the fold | Same table below the fold |
Evidence by reference — every quote is attributed to the persona that said it.
The outcome-led hero ('Bank in an afternoon, not three weeks') converted first-time and international founders who read the headline as an eligibility promise.
Cost-scrutinising and finance personas ignored both heroes and scrolled to the table — the hero is not where they decide.
Risk-averse founders trusted the current hero more; the new one read as 'too fast to be safe'.
Each persona ran hundreds of independent users. Click a persona to scope the whole report.
“Either. I scrolled to spend controls.”
“I don't read headlines. I read the fee rows.”
“'In an afternoon' told me I wouldn't be sent to a branch.”
| Persona | Split | A | B | None | Deciding signal |
|---|---|---|---|---|---|
| ASAI-heavy spender 200 users | 44.0% | 52.0% | 4.0% | Indifferent | |
| CFCost-scrutinising founder 200 users | 46.0% | 48.0% | 6.0% | Indifferent | |
| PFFirst-time US founder 200 users | 20.5% | 78.0% | 1.5% | Eligibility promise | |
| PSSolo technical founder 200 users | 45.0% | 50.5% | 4.5% | Indifferent | |
| RFRisk-averse founder 200 users | 60.5% | 33.0% | 6.5% | Trust tone | |
| RBRevenue-funded business owner 200 users | 42.0% | 54.0% | 4.0% | Plain language | |
| SFSupport-seeking founder 200 users | 29.0% | 66.5% | 4.5% | Clear next step | |
| SWSwitching founder 200 users | 48.5% | 44.5% | 7.0% | Indifferent | |
| NFNon-US founder applying remotely 200 users | 19.5% | 79.0% | 1.5% | Remote applicant fit | |
| FOFinance operator on a small team 200 users | 47.5% | 46.5% | 6.0% | Indifferent | |
| MFMobile-first founder 200 users | 42.0% | 53.0% | 5.0% | Mobile hero legibility | |
| BKBookkeeper / fractional CFO 200 users | 48.5% | 47.0% | 4.5% | Indifferent |
What the simulated users said about each variant, in their own words.
Ordered by impact. Fixes that apply to both variants come first.
Cost-focused personas never engage with the hero; bringing one fee fact up captures them earlier.
Recovers the risk-averse personas without losing the speed message.
What the stronger variant gives up that the other does well.
Risk-averse persona preference moved 9 points toward A.
How this test read on earlier runs.
How participants answered each question.
“That I could finish today and wouldn't need a branch visit.”
Per-step reaction replay — what every persona thought at every step.
#12,400 simulated users each judged both variants in a live browser. They are not real visitors, so treat every number as a direction to check rather than a measurement. Selection share is a count-based rate recomputed on read; A/B is between-subjects at the respondent layer and paired at the driver layer, so the preference counts are reported beside conversion, never mixed into it. Judging engine version judge-2.4 · accuracy benchmark pending.