## Landing Page CTAs
- Above fold: Primary action
- After benefits: Reinforcement
- After testimonials: Social proof boost
- Multiple times without overwhelming
```

**What to test, and in what order.** Run ONE change per test (an "element" test), not a redesign, or you can't attribute the lift. Sequence tests by expected effect size — biggest lever first:

| Priority | Element | What to vary | Why first/last |
|----------|---------|--------------|----------------|
| 1 | Copy / message | "Get Started Free" vs "Start Your Free Trial" vs "Try It Risk-Free" vs "Book a Demo" | Message-match drives the largest swings; cheapest to change |
| 2 | Offer & friction | Free trial vs demo, credit-card vs no-card, form-field count | Changing the *ask* moves intent more than styling |
| 3 | Placement | Hero only, hero + sticky footer, repeat after social proof, exit-intent | Visibility/repetition compounds; test after message is settled |
| 4 | Visual | Color/contrast, size, filled vs outlined | Real but usually small; only worth it on high-traffic pages |
| 5 | Urgency (handle with care) | "Ends Friday", "3 seats left" — only if the scarcity is *real* | Fake countdowns/scarcity are deceptive and FTC-actionable (see Claims & Compliance) |

Don't call a winner on raw conversion-rate difference. Pre-compute the sample size you need, run to it, then check significance — see [Copy Testing & Optimization](#copy-testing--optimization) for the formulas.

---