Brazil receives its own Portuguese funnel. Two hero variants are compared with the same offer, attribution rules and period so the result is not caused by configuration differences.
Actions
- Validate the form, currency display and payment step
- Send equal test volume to both versions
- Compare FTD rate rather than clicks alone
- Keep the winner after cohort-quality review
Situation
Portuguese localisation, mobile creative, permitted sources, commercial terms and cohort monitoring for a Brazil launch.
Red flags
- Use a machine-translated Spanish page
- Change creative and audience at the same time
- Declare a winner from a tiny registration sample
Why appearance does not choose the winner
Both versions use equivalent audiences and the same attribution rules. The winner is selected through FTD and early redeposit rather than clicks or visual preference. Changing creative and targeting at the same time would make the test inconclusive.
A clean two-funnel test
Give both versions comparable audiences, equal budgets and separate links. Change one meaningful element, such as first-screen wording or mobile-form length. If creative, targeting and landing page all change together, the result cannot explain which decision caused the difference.
Select the winner through FTD and early redeposit. CTR remains useful for diagnosis, but a visually attractive version can still produce inexpensive and low-value registrations.
BR is not a translated Spanish-LATAM campaign
Brazilian Portuguese differs visibly from Spanish LATAM assets and from European Portuguese. Check copy, currency formatting, mobile layout and the deposit journey. Casino and sports should use separate campaigns because their intent and post-registration behaviour differ.
When scaling, measure quality after FTD. A strong first deposit does not compensate for weak later activity when the payment model depends on longer-term NGR.
The checklist produces a localised, measurable funnel. Scale follows FTD and early redeposit, not visual preference.
Frequently asked questions
How do you pick a winner between two Brazil funnels?
By FTD and early redeposit, not CTR or click volume — a high click-through rate can produce cheap but weak registrations.
Why shouldn't you change the creative and targeting at the same time in an A/B test?
You won't be able to tell what actually drove the result — versions should differ by only one testable element, with equal budget and audience.
Is translating a Spanish landing page into Portuguese enough for Brazil?
No, you need material written for Brazil from the start, accounting for local payment behaviour — not a machine translation from Spanish.