Skills › Marketing Skills › ab-testing
ab-testing
The facts
Its description
When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program," or "experiment playbook." Use this whenever someone is comparing two approaches and wants to measure which performs better, or when they want to build a systematic experimentation practice. For tracking implementation, see analytics. For page-level conversion optimization, see cro.
The description the agent reads to decide when to use it.
From its settingsline 3
When it starts
The agent picks this skill when someone wants to plan, design, run, or interpret an A/B test or experiment, or build an ongoing growth experimentation program — essentially any time two approaches are being compared and the user wants to know which performs better. The description is broad and trigger-heavy, so it fires on test-design, sample-size, significance, and experimentation-program questions.
Things a person might say (the model's guesses)
- “Should I A/B test a new headline on our pricing page?”
- “How long do I need to run this test before the results are trustworthy?”
- “We want to build a real growth experimentation program — where do we start?”
A model's summary of when it is used.
A model's reading
Files it mentions
references/sample-size-guide.md, references/test-templates.md
Files the text names, such as ones to read or write. Found by matching words, so treat them as pointers.
What it leads to
- requires product-marketing when a product marketing context file exists line 15
Other skills this one hands off to, calls, requires or offers instead, as a model read them.
A model's reading
What leads to it
- marketing-council hands off to this when the disagreement map says test it marketing-council, line 160
- ads calls this for landing-page test design ads, line 322
- churn-prevention calls this when running cancel flow experiments churn-prevention, line 373
- copywriting calls this copywriting, line 339
- cro calls this when experiment ideas need test design and measurement cro, line 19
- marketing-loops calls this marketing-loops/loop-catalog.md, line 520
- offers calls this offers, line 114
- signup calls this when testing signup flow changes signup, line 363
- analytics is an alternative to this for A/B test measurement analytics, line 3
Other skills that hand off to this one, call it, require it or offer it instead, as a model read them.
A model's reading
Licence
MIT, from LICENSE
The terms its author shares it under, which let it be shown here.
From the repositorylicence text
Other settings
metadata: version: 2.0.4
More settings at the top of the file, shown as written.
From its settingsline 4
Other files in this skill (3) Close
The text
Most marks come from matching words in the file, so treat them as pointers.
1---
2name: ab-testing
3description: When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program," or "experiment playbook." Use this whenever someone is comparing two approaches and wants to measure which performs better, or when they want to build a systematic experimentation practice. For tracking implementation, see analytics. For page-level conversion optimization, see cro.
4metadata:
5 version: 2.0.4
6---
8# A/B Test Setup
10You are an expert in experimentation and A/B testing. Your goal is to help design tests that produce statistically valid, actionable results.
A/B Test Setup
You are an expert in experimentation and A/B testing. Your goal is to help design tests that produce statistically valid, actionable results.
Raw lines 1 to 1112## Initial Assessment
14**Check for product marketing context first:**
15If `.agents/product-marketing.md` exists (or `.claude/product-marketing.md`, or the legacy `product-marketing-context.md` filename, in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
17Before designing a test, understand:
191. **Test Context** - What are you trying to improve? What change are you considering?
202. **Current State** - Baseline conversion rate? Current traffic volume?
213. **Constraints** - Technical complexity? Timeline? Tools available?
23---
Initial Assessment
Check for product marketing context first: If .agents/product-marketing.md exists (or .claude/product-marketing.md, or the legacy product-marketing-context.md filename, in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
Before designing a test, understand:
- Test Context - What are you trying to improve? What change are you considering?
- Current State - Baseline conversion rate? Current traffic volume?
- Constraints - Technical complexity? Timeline? Tools available?
Raw lines 12 to 24
Core Principles
Raw lines 25 to 2627### 1. Start with a Hypothesis
28- Not just "let's see what happens"
29- Specific prediction of outcome
30- Based on reasoning or data
1. Start with a Hypothesis
- Not just "let's see what happens"
- Specific prediction of outcome
- Based on reasoning or data
36### 3. Statistical Rigor
37- Pre-determine sample size
38- Don't peek and stop early
39- Commit to the methodology
3. Statistical Rigor
- Pre-determine sample size
- Don't peek and stop early
- Commit to the methodology
41### 4. Measure What Matters
42- Primary metric tied to business value
43- Secondary metrics for context
44- Guardrail metrics to prevent harm
46---
4. Measure What Matters
- Primary metric tied to business value
- Secondary metrics for context
- Guardrail metrics to prevent harm
Raw lines 41 to 47
Hypothesis Framework
Raw lines 48 to 4950### Structure
52```
53Because [observation/data],
54we believe [change]
55will cause [expected outcome]
56for [audience].
57We'll know this is true when [metrics].
58```
Structure
Because [observation/data],
we believe [change]
will cause [expected outcome]
for [audience].
We'll know this is true when [metrics].60### Example
62**Weak**: "Changing the button color might increase clicks."
64**Strong**: "Because users report difficulty finding the CTA (per heatmaps and feedback), we believe making the button larger and using contrasting color will increase CTA clicks by 15%+ for new visitors. We'll measure click-through rate from page view to signup start."
66---
Example
Weak: "Changing the button color might increase clicks."
Strong: "Because users report difficulty finding the CTA (per heatmaps and feedback), we believe making the button larger and using contrasting color will increase CTA clicks by 15%+ for new visitors. We'll measure click-through rate from page view to signup start."
Raw lines 60 to 67
68## Test Types
70| Type | Description | Traffic Needed |
71|------|-------------|----------------|
72| A/B | Two versions, single change | Moderate |
73| A/B/n | Multiple variants | Higher |
74| MVT | Multiple changes in combinations | Very high |
75| Split URL | Different URLs for variants | Moderate |
77---
Test Types
| Type | Description | Traffic Needed |
|---|---|---|
| A/B | Two versions, single change | Moderate |
| A/B/n | Multiple variants | Higher |
| MVT | Multiple changes in combinations | Very high |
| Split URL | Different URLs for variants | Moderate |
Raw lines 68 to 78
Sample Size
Raw lines 79 to 8081### Quick Reference
83| Baseline | 10% Lift | 20% Lift | 50% Lift |
84|----------|----------|----------|----------|
85| 1% | 160k/variant | 43k/variant | 7.7k/variant |
86| 3% | 53k/variant | 14k/variant | 2.5k/variant |
87| 5% | 31k/variant | 8.2k/variant | 1.5k/variant |
88| 10% | 15k/variant | 3.8k/variant | 690/variant |
90Per variant, for 95% confidence (two-sided) and 80% power. Lift is relative (3% → 3.6% is a 20% lift).
92**Calculators:**
93- [Evan Miller's](https://www.evanmiller.org/ab-testing/sample-size.html)
94- [Optimizely's](https://www.optimizely.com/sample-size-calculator/)
96**For detailed sample size tables and duration calculations**: See [references/sample-size-guide.md](references/sample-size-guide.md)
98---
Quick Reference
| Baseline | 10% Lift | 20% Lift | 50% Lift |
|---|---|---|---|
| 1% | 160k/variant | 43k/variant | 7.7k/variant |
| 3% | 53k/variant | 14k/variant | 2.5k/variant |
| 5% | 31k/variant | 8.2k/variant | 1.5k/variant |
| 10% | 15k/variant | 3.8k/variant | 690/variant |
Per variant, for 95% confidence (two-sided) and 80% power. Lift is relative (3% → 3.6% is a 20% lift).
Calculators:
For detailed sample size tables and duration calculations: See references/sample-size-guide.md
Raw lines 81 to 99
Metrics Selection
Raw lines 100 to 101102### Primary Metric
103- Single metric that matters most
104- Directly tied to hypothesis
105- What you'll use to call the test
Primary Metric
- Single metric that matters most
- Directly tied to hypothesis
- What you'll use to call the test
107### Secondary Metrics
108- Support primary metric interpretation
109- Explain why/how the change worked
Secondary Metrics
- Support primary metric interpretation
- Explain why/how the change worked
111### Guardrail Metrics
112- Things that shouldn't get worse
113- Stop test if significantly negative
Guardrail Metrics
- Things that shouldn't get worse
- Stop test if significantly negative
115### Example: Pricing Page Test
116- **Primary**: Plan selection rate
117- **Secondary**: Time on page, plan distribution
118- **Guardrail**: Support tickets, refund rate
Example: Pricing Page Test
- Primary: Plan selection rate
- Secondary: Time on page, plan distribution
- Guardrail: Support tickets, refund rate
120### Opt-In Elements: Measure the Exposed Group
121When a change only reaches visitors who interact with it (a hero video, a calculator, a chat widget), a test on overall conversion is underpowered: if 10% of visitors click play, the other 90% add noise and no signal. Either run a triggered analysis (log the same trigger point in both arms, such as the play button coming into view, and compare only visitors who reached it; don't filter on clicks, since the variant can change who clicks), or skip the test, ship the better version, and compare versions on engagement (watch-through) and viewer conversion, tagged by version ID. The second is a directional comparison, not a controlled test.
123---
Opt-In Elements: Measure the Exposed Group
When a change only reaches visitors who interact with it (a hero video, a calculator, a chat widget), a test on overall conversion is underpowered: if 10% of visitors click play, the other 90% add noise and no signal. Either run a triggered analysis (log the same trigger point in both arms, such as the play button coming into view, and compare only visitors who reached it; don't filter on clicks, since the variant can change who clicks), or skip the test, ship the better version, and compare versions on engagement (watch-through) and viewer conversion, tagged by version ID. The second is a directional comparison, not a controlled test.
Raw lines 120 to 124
Designing Variants
Raw lines 125 to 126127### What to Vary
129| Category | Examples |
130|----------|----------|
131| Headlines/Copy | Message angle, value prop, specificity, tone |
132| Visual Design | Layout, color, images, hierarchy |
133| CTA | Button copy, size, placement, number |
134| Content | Information included, order, amount, social proof |
What to Vary
| Category | Examples |
|---|---|
| Headlines/Copy | Message angle, value prop, specificity, tone |
| Visual Design | Layout, color, images, hierarchy |
| CTA | Button copy, size, placement, number |
| Content | Information included, order, amount, social proof |
136### Best Practices
137- Single, meaningful change
138- Bold enough to make a difference
139- True to the hypothesis
141---
Best Practices
- Single, meaningful change
- Bold enough to make a difference
- True to the hypothesis
Raw lines 136 to 142
143## Traffic Allocation
145| Approach | Split | When to Use |
146|----------|-------|-------------|
147| Standard | 50/50 | Default for A/B |
148| Conservative | 90/10, 80/20 | Limit risk of bad variant |
149| Ramping | Start small, increase | Technical risk mitigation |
151**Considerations:**
152- Consistency: Users see same variant on return
153- Balanced exposure across time of day/week
155---
Traffic Allocation
| Approach | Split | When to Use |
|---|---|---|
| Standard | 50/50 | Default for A/B |
| Conservative | 90/10, 80/20 | Limit risk of bad variant |
| Ramping | Start small, increase | Technical risk mitigation |
Considerations:
- Consistency: Users see same variant on return
- Balanced exposure across time of day/week
Raw lines 143 to 156
Implementation
Raw lines 157 to 158159### Client-Side
160- JavaScript modifies page after load
161- Quick to implement, can cause flicker
162- Tools: PostHog, Optimizely, VWO
Client-Side
- JavaScript modifies page after load
- Quick to implement, can cause flicker
- Tools: PostHog, Optimizely, VWO
164### Server-Side
165- Variant determined before render
166- No flicker, requires dev work
167- Tools: PostHog, LaunchDarkly, Split
169---
Server-Side
- Variant determined before render
- No flicker, requires dev work
- Tools: PostHog, LaunchDarkly, Split
Raw lines 164 to 170
Running the Test
Raw lines 171 to 172173### Pre-Launch Checklist
174- [ ] Hypothesis documented
175- [ ] Primary metric defined
176- [ ] Sample size calculated
177- [ ] Variants implemented correctly
178- [ ] Tracking verified
179- [ ] QA completed on all variants
Pre-Launch Checklist
- [ ] Hypothesis documented
- [ ] Primary metric defined
- [ ] Sample size calculated
- [ ] Variants implemented correctly
- [ ] Tracking verified
- [ ] QA completed on all variants
181### During the Test
183**DO:**
184- Monitor for technical issues
185- Check segment quality
186- Document external factors
188**Avoid:**
189- Peek at results and stop early
190- Make changes to variants
191- Add traffic from new sources
During the Test
DO:
- Monitor for technical issues
- Check segment quality
- Document external factors
Avoid:
- Peek at results and stop early
- Make changes to variants
- Add traffic from new sources
193### The Peeking Problem
194Looking at results before reaching sample size and stopping early leads to false positives and wrong decisions. Pre-commit to sample size and trust the process.
196---
The Peeking Problem
Looking at results before reaching sample size and stopping early leads to false positives and wrong decisions. Pre-commit to sample size and trust the process.
Raw lines 193 to 197
Analyzing Results
Raw lines 198 to 199200### Statistical Significance
201- For a pre-specified fixed-horizon test, compare its p-value with the chosen significance level (commonly α = 0.05).
202- A p-value describes how incompatible the data are with the null model: the probability, **assuming that model and its assumptions**, of a result at least as extreme as the observed one. It is not the probability that the result is random, that the null is true, or that a variant will win again.
203- A 95% confidence interval comes from a procedure with 95% coverage over repeated samples under its assumptions. It is not a 95% posterior probability that this particular interval contains the true effect. Report the interval and effect size alongside the p-value.
204- Use the planned analysis and stopping rule. Sequential tests and Bayesian analyses have their own interpretation; do not translate every platform's “confidence” score into a fixed-horizon p-value.
206These distinctions follow the [American Statistical Association's p-value statement](https://www.amstat.org/asa/files/pdfs/p-valuestatement.pdf). Statistical significance alone does not establish business value or justify shipping a variant.
Statistical Significance
- For a pre-specified fixed-horizon test, compare its p-value with the chosen significance level (commonly α = 0.05).
- A p-value describes how incompatible the data are with the null model: the probability, assuming that model and its assumptions, of a result at least as extreme as the observed one. It is not the probability that the result is random, that the null is true, or that a variant will win again.
- A 95% confidence interval comes from a procedure with 95% coverage over repeated samples under its assumptions. It is not a 95% posterior probability that this particular interval contains the true effect. Report the interval and effect size alongside the p-value.
- Use the planned analysis and stopping rule. Sequential tests and Bayesian analyses have their own interpretation; do not translate every platform's “confidence” score into a fixed-horizon p-value.
These distinctions follow the American Statistical Association's p-value statement. Statistical significance alone does not establish business value or justify shipping a variant.
Raw lines 200 to 207208### Analysis Checklist
2101. **Reach sample size?** If not, result is preliminary
2112. **Statistically significant?** Check confidence intervals
2123. **Effect size meaningful?** Compare to MDE, project impact
2134. **Secondary metrics consistent?** Support the primary?
2145. **Guardrail concerns?** Anything get worse?
2156. **Segment differences?** Mobile vs. desktop? New vs. returning?
Analysis Checklist
- Reach sample size? If not, result is preliminary
- Statistically significant? Check confidence intervals
- Effect size meaningful? Compare to MDE, project impact
- Secondary metrics consistent? Support the primary?
- Guardrail concerns? Anything get worse?
- Segment differences? Mobile vs. desktop? New vs. returning?
217### Interpreting Results
219| Result | Conclusion |
220|--------|------------|
221| Significant winner | Check effect size, uncertainty and guardrails before implementing |
222| Significant loser | Keep control, learn why |
223| No significant difference | Report the effect interval: it may still allow meaningful benefit and harm, or may exclude effects worth pursuing. Follow the planned stopping rule; do not extend a finished fixed-horizon test until it becomes significant. Plan a new test if needed. |
224| Mixed signals | Check data quality and planned metrics; label unplanned segment findings exploratory and confirm them in a new test |
226---
Interpreting Results
| Result | Conclusion |
|---|---|
| Significant winner | Check effect size, uncertainty and guardrails before implementing |
| Significant loser | Keep control, learn why |
| No significant difference | Report the effect interval: it may still allow meaningful benefit and harm, or may exclude effects worth pursuing. Follow the planned stopping rule; do not extend a finished fixed-horizon test until it becomes significant. Plan a new test if needed. |
| Mixed signals | Check data quality and planned metrics; label unplanned segment findings exploratory and confirm them in a new test |
Raw lines 217 to 227
228## Documentation
230Document every test with:
231- Hypothesis
232- Variants (with screenshots)
233- Results (sample, metrics, significance)
234- Decision and learnings
236**For templates**: See [references/test-templates.md](references/test-templates.md)
238---
Documentation
Document every test with:
- Hypothesis
- Variants (with screenshots)
- Results (sample, metrics, significance)
- Decision and learnings
For templates: See references/test-templates.md
Raw lines 228 to 239
240## Growth Experimentation Program
242Individual tests are valuable. A continuous experimentation program is a compounding asset. This section covers how to run experiments as an ongoing growth engine, not just one-off tests.
Growth Experimentation Program
Individual tests are valuable. A continuous experimentation program is a compounding asset. This section covers how to run experiments as an ongoing growth engine, not just one-off tests.
Raw lines 240 to 243244### The Experiment Loop
246```
2471. Generate hypotheses (from data, research, competitors, customer feedback)
2482. Prioritize with ICE scoring
2493. Design and run the test
2504. Analyze results with statistical rigor
2515. Promote winners to a playbook
2526. Generate new hypotheses from learnings
253→ Repeat
254```
The Experiment Loop
1. Generate hypotheses (from data, research, competitors, customer feedback)
2. Prioritize with ICE scoring
3. Design and run the test
4. Analyze results with statistical rigor
5. Promote winners to a playbook
6. Generate new hypotheses from learnings
→ Repeat256### Hypothesis Generation
258Feed your experiment backlog from multiple sources:
260| Source | What to Look For |
261|--------|-----------------|
262| Analytics | Drop-off points, low-converting pages, underperforming segments |
263| Customer research | Pain points, confusion, unmet expectations |
264| Competitor analysis | Features, messaging, or UX patterns they use that you don't |
265| Support tickets | Recurring questions or complaints about conversion flows |
266| Heatmaps/recordings | Where users hesitate, rage-click, or abandon |
267| Past experiments | "Significant loser" tests often reveal new angles to try |
Hypothesis Generation
Feed your experiment backlog from multiple sources:
| Source | What to Look For |
|---|---|
| Analytics | Drop-off points, low-converting pages, underperforming segments |
| Customer research | Pain points, confusion, unmet expectations |
| Competitor analysis | Features, messaging, or UX patterns they use that you don't |
| Support tickets | Recurring questions or complaints about conversion flows |
| Heatmaps/recordings | Where users hesitate, rage-click, or abandon |
| Past experiments | "Significant loser" tests often reveal new angles to try |
269### ICE Prioritization
271Score each hypothesis 1-10 on three dimensions:
273| Dimension | Question |
274|-----------|----------|
275| **Impact** | If this works, how much will it move the primary metric? |
276| **Confidence** | How sure are we this will work? (Based on data, not gut.) |
277| **Ease** | How fast and cheap can we ship and measure this? |
279**ICE Score** = (Impact + Confidence + Ease) / 3
281Run highest-scoring experiments first. Re-score monthly as context changes.
ICE Prioritization
Score each hypothesis 1-10 on three dimensions:
| Dimension | Question |
|---|---|
| Impact | If this works, how much will it move the primary metric? |
| Confidence | How sure are we this will work? (Based on data, not gut.) |
| Ease | How fast and cheap can we ship and measure this? |
ICE Score = (Impact + Confidence + Ease) / 3
Run highest-scoring experiments first. Re-score monthly as context changes.
Raw lines 269 to 282283### Experiment Velocity
285Track your experimentation rate as a leading indicator of growth:
287| Metric | Target |
288|--------|--------|
289| Experiments launched per month | 4-8 for most teams |
290| Win rate | 20-30% is common for mature programs (sustained higher rates may indicate conservative hypotheses) |
291| Average test duration | 2-4 weeks |
292| Backlog depth | 20+ hypotheses queued |
293| Cumulative lift | Compound gains from all winners |
Experiment Velocity
Track your experimentation rate as a leading indicator of growth:
| Metric | Target |
|---|---|
| Experiments launched per month | 4-8 for most teams |
| Win rate | 20-30% is common for mature programs (sustained higher rates may indicate conservative hypotheses) |
| Average test duration | 2-4 weeks |
| Backlog depth | 20+ hypotheses queued |
| Cumulative lift | Compound gains from all winners |
295### The Experiment Playbook
297When a test wins, don't just implement it — document the pattern:
299```
300## [Experiment Name]
301**Date**: [date]
302**Hypothesis**: [the hypothesis]
303**Sample size**: [n per variant]
304**Result**: [winner/loser/inconclusive] — [primary metric] changed by [X%] (95% CI: [range], p=[value])
305**Guardrails**: [any guardrail metrics and their outcomes]
306**Segment deltas**: [notable differences by device, segment, or cohort]
307**Why it worked/failed**: [analysis]
308**Pattern**: [the reusable insight — e.g., "social proof near pricing CTAs increases plan selection"]
309**Apply to**: [other pages/flows where this pattern might work]
310**Status**: [implemented / parked / needs follow-up test]
311```
313Over time, your playbook becomes a library of proven growth patterns specific to your product and audience.
The Experiment Playbook
When a test wins, don't just implement it — document the pattern:
## [Experiment Name]
**Date**: [date]
**Hypothesis**: [the hypothesis]
**Sample size**: [n per variant]
**Result**: [winner/loser/inconclusive] — [primary metric] changed by [X%] (95% CI: [range], p=[value])
**Guardrails**: [any guardrail metrics and their outcomes]
**Segment deltas**: [notable differences by device, segment, or cohort]
**Why it worked/failed**: [analysis]
**Pattern**: [the reusable insight — e.g., "social proof near pricing CTAs increases plan selection"]
**Apply to**: [other pages/flows where this pattern might work]
**Status**: [implemented / parked / needs follow-up test]Over time, your playbook becomes a library of proven growth patterns specific to your product and audience.
Raw lines 295 to 314315### Experiment Cadence
317**Weekly (30 min)**: Review running experiments for technical issues and guardrail metrics. Don't call winners early — but do stop tests where guardrails are significantly negative.
319**Bi-weekly**: Conclude completed experiments. Analyze results, update playbook, launch next experiment from backlog.
321**Monthly (1 hour)**: Review experiment velocity, win rate, cumulative lift. Replenish hypothesis backlog. Re-prioritize with ICE.
323**Quarterly**: Audit the playbook. Which patterns have been applied broadly? Which winning patterns haven't been scaled yet? What areas of the funnel are under-tested?
325---
Experiment Cadence
Weekly (30 min): Review running experiments for technical issues and guardrail metrics. Don't call winners early — but do stop tests where guardrails are significantly negative.
Bi-weekly: Conclude completed experiments. Analyze results, update playbook, launch next experiment from backlog.
Monthly (1 hour): Review experiment velocity, win rate, cumulative lift. Replenish hypothesis backlog. Re-prioritize with ICE.
Quarterly: Audit the playbook. Which patterns have been applied broadly? Which winning patterns haven't been scaled yet? What areas of the funnel are under-tested?
Raw lines 315 to 326
Common Mistakes
Raw lines 327 to 328329### Test Design
330- Testing too small a change (undetectable)
331- Testing too many things (can't isolate)
332- No clear hypothesis
Test Design
- Testing too small a change (undetectable)
- Testing too many things (can't isolate)
- No clear hypothesis
339### Analysis
340- Ignoring confidence intervals
341- Cherry-picking segments
342- Over-interpreting inconclusive results
344---
Analysis
- Ignoring confidence intervals
- Cherry-picking segments
- Over-interpreting inconclusive results
Raw lines 339 to 345
346## Task-Specific Questions
3481. What's your current conversion rate?
3492. How much traffic does this page get?
3503. What change are you considering and why?
3514. What's the smallest improvement worth detecting?
3525. What tools do you have for testing?
3536. Have you tested this area before?
355---
Task-Specific Questions
- What's your current conversion rate?
- How much traffic does this page get?
- What change are you considering and why?
- What's the smallest improvement worth detecting?
- What tools do you have for testing?
- Have you tested this area before?
Raw lines 346 to 356
357## Related Skills
361- **copywriting**: For creating variant copy
Related Skills
- cro: For generating test ideas based on CRO principles
- analytics: For setting up test measurement
- copywriting: For creating variant copy
From coreyhaines31/marketingskills at 88c79f6, 2026-10-30. Skills read by DeepSeek V4.1 Flash on 2026-11-04. Summary and techniques by Claude Opus 5.5 on 2026-11-06.