ranbot-ai/ab-test-setup

A/B Test Setup

1️⃣ Purpose & Scope

Ensure every A/B test is valid, rigorous, and safe before a single line of code is written.

• Prevents "peeking"
• Enforces statistical power
• Blocks invalid hypotheses

---

2️⃣ Pre-Requisites

You must have:

• A clear user problem
• Access to an analytics source
• Roughly estimated traffic volume

Hypothesis Quality Checklist

A valid hypothesis includes:

• Observation or evidence
• Single, specific change
• Directional expectation
• Defined audience
• Measurable success criteria

---

3️⃣ Hypothesis Lock (Hard Gate)

Before designing variants or metrics, you MUST:

• Present the final hypothesis
• Specify:
  • Target audience
  • Primary metric
  • Expected direction of effect
  • Minimum Detectable Effect (MDE)

Ask explicitly:

“Is this the final hypothesis we are committing to for this test?”

Do NOT proceed until confirmed.

---

4️⃣ Assumptions & Validity Check (Mandatory)

Explicitly list assumptions about:

• Traffic stability
• User independence
• Metric reliability
• Randomization quality
• External factors (seasonality, campaigns, releases)

If assumptions are weak or violated:

• Warn the user
• Recommend delaying or redesigning the test

---

5️⃣ Test Type Selection

Choose the simplest valid test:

• A/B Test – single change, two variants
• A/B/n Test – multiple variants, higher traffic required
• Multivariate Test (MVT) – interaction effects, very high traffic
• Split URL Test – major structural changes

Default to A/B unless there is a clear reason otherwise.

---

6️⃣ Metrics Definition

Primary Metric (Mandatory)

• Single metric used to evaluate success
• Directly tied to the hypothesis
• Pre-defined and frozen before launch

Secondary Metrics

• Provide context
• Explain why results occurred
• Must not override the primary metric

Guardrail Metrics

• Metrics that must not degrade
• Used to prevent harmful wins
• Trigger test stop if significantly negative

---

7️⃣ Sample Size & Duration

Define upfront:

• Baseline rate
• MDE
• Significance level (typically 95%)
• Statistical power (typically 80%)

Estimate:

• Required sample size per variant
• Expected test duration

Do NOT proceed without a realistic sample size estimate.

---

Tracking Verification (Required before Gate 8)

Before entering the Execution Readiness Gate below, run through this checklist to make "Tracking is verified" mean something concrete:

1. Event firing: Trigger each event the primary and secondary metrics depend on (sign-up, add-to-cart, custom event) on staging or a debug page, and confirm it lands in your analytics destination within 30 seconds.
2. Variant attribution: Verify that the variant assignment ID is attached to every fired event — not just the entry event. Use your analytics' raw event view to compare a sample of 5+ events per variant.
3. De-duplication: Confirm that a user reloading the page does not cause double-counted events. If your stack uses client-side de-duping, the variant ID must be part of the dedup key.
4. Sample randomization: Pull the first 100 assignment records from your assignment table; the variant split should be within ±5% of the configured allocation.
5. Guardrail metric pipeline: Each guardrail metric defined in §6️⃣ must have a working dashboard or alert by the time the test launches.

If any of the above fails, stop and resolve it before Gate 8.

---

8️⃣ Execution Readiness Gate (Hard Stop)

You may proceed to implementation only if all are true:

• Hypothesis is locked
• Primary metric is frozen
• Sample size is calculated
• Test duration is defined
• Guardrails are set
• Tracking is verified

If any item is missing, stop and resolve it.

---

Running the Test

During the Test

DO:

• Monitor technical health
• Document external factors

DO NOT:

• Stop early due to “good-looking” results
• Change variants mid-test
• Add new traffic sources
• Redefine success criteria

---

Analyzing Results

Analysis Discipline

When interpreting results:

• Do NOT generalize beyond the tested population
• Do NOT claim causality beyond the tested change
• Do NOT override guardrail failures
• Separate statistical significance from business judgment

Interpretation Outcomes

| Result | Action | | -------------------- | -------------------------------------- | | Significant positive | Consider rollout | | Significant negative | Reject variant, document learning | | Inconclusive | Consider more traffic or bolder change | | Guardrail failure | Do not ship, even if primary wins |

---

Documentation & Learning

Test Record (Mandatory)

Document:

• Hypothesis
• Variants
• Metrics
• Sample size vs achieved
• Results
• Decision
• Learnings
• Follow-up ideas

Store records in a shared, searchable location to avoid repeated