# Ad Creative Test Log — Quality Checklist and Tracking Template

## Part 1 — Test Design Quality Checklist

Use before launching any A/B test to verify the design is valid.

### Category 1 — Hypothesis Quality (5 items)
- [ ] Hypothesis written before test started (not after seeing results)
- [ ] Hypothesis follows "If / Then / Because" structure
- [ ] Predicted outcome includes a specific number (e.g., "15% CPA reduction", not "better performance")
- [ ] "Because" section references evidence or a specific reasoning (not "because we think so")
- [ ] Single variable identified clearly

**Score — __ / 5**

### Category 2 — Test Structure (6 items)
- [ ] Only one variable differs between variants
- [ ] All other elements confirmed identical (offer, copy, CTA, format, audience, budget)
- [ ] 50/50 budget split (or asymmetric split documented with rationale)
- [ ] Platform automatic creative optimization disabled during test
- [ ] Audience exclusions in place if running audience segment test
- [ ] Campaign/ad set structure supports clean data isolation

**Score — __ / 6**

### Category 3 — Sample Size and Duration (5 items)
- [ ] Minimum sample size calculated before launch (using significance calculator)
- [ ] Sample size is ≥200 conversions per variant (minimum) or ≥500 (preferred)
- [ ] Test duration is at least 7 days
- [ ] Budget per variant sufficient to reach sample size within planned duration
- [ ] Start and end dates documented

**Score — __ / 5**

### Category 4 — Metrics and Decision Criteria (5 items)
- [ ] Primary metric selected before test starts
- [ ] Primary metric is a business outcome (CPA, ROAS, conversion rate) — not a vanity metric
- [ ] Statistical confidence threshold set (95% preferred; 90% minimum)
- [ ] Clear winner declaration criteria defined (minimum improvement threshold)
- [ ] Early kill criteria defined (e.g., CPA 3× baseline after 100 conversions)

**Score — __ / 5**

**Total pre-launch score — __ / 21**
- **19–21**: Good to launch
- **15–18**: Address flagged items before launching
- **Below 15**: Redesign the test

---

## Part 2 — Test Results Log

Use this table to record each test result. Build this log over time — it becomes a compound advantage.

| # | Date | Product | Platform | Variable Tested | Control (A) | Challenger (B) | Primary Metric (A) | Primary Metric (B) | Improvement | Confidence | Winner | Key Learning |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 001 | [Date] | [Product] | [Platform] | [Hook/Offer/LP/etc.] | [Description] | [Description] | [Value] | [Value] | [+/-x%] | [x%] | A/B/None | [1 sentence] |
| 002 | | | | | | | | | | | | |
| 003 | | | | | | | | | | | | |

---

## Part 3 — Learning Summary by Variable Type

Summarize patterns from your test history to guide future test design.

### Hook Learnings
- [Date]: [What hook type won and by how much for what product/audience]
- [Date]: [Pattern or insight from hook tests]

### Offer Learnings
- [Date]: [What offer structure won and by how much]

### Landing Page Learnings
- [Date]: [What page element change had the biggest impact]

### Audience Learnings
- [Date]: [What audience finding applied broadly]

### Format Learnings
- [Date]: [Format test results — video vs. static, short vs. long, etc.]

---

## Part 4 — Creative Brief Implications

Patterns from tests that should inform your default creative brief:

| Finding | Current Default | New Default Informed by Testing |
|---|---|---|
| Hook type | [e.g., Product demonstration] | [e.g., Pain-point question — 22% lower CPA across 3 tests] |
| Video length | [e.g., 30 seconds] | [e.g., 15 seconds — equal conversion, lower CPM] |
| Offer frame | [e.g., Percentage discount] | [e.g., Bonus gift — better margin and equivalent conversion] |
| CTA text | [e.g., "Shop Now"] | [e.g., "Get Yours" — 8% higher CTR in 2 tests] |

Update this table after every 3–5 tests. Your creative brief should evolve based on actual test data, not intuition.