mxdify — Growth infrastructure for digital health and SaaS
Stage 04 of 07 · Closed-Loop Growth System

Experiment Design

Experiment design is the discipline of translating a hypothesis into an experimental protocol that will produce a readable result. It specifies the unit of randomization, the primary and guardrail metrics, the required sample size, the exposure duration, and the analysis plan before the experiment is launched.

Written by Andrew Eastlick·Published
Key takeaways
  • The experiment is designed to be readable. If a null result cannot be trusted, the design is wrong.
  • Sample size is calculated against the minimum detectable effect that would be operationally meaningful, not against convenience.
  • Instrumentation for the primary metric is verified end-to-end before allocation is opened.
  • The analysis plan is pre-registered. Post-hoc metric shopping is not allowed.
The problem it solves

Where does this stage earn its keep?

Underpowered tests, contaminated allocations, and metrics that were not instrumented for the surface being changed. The result is months of experiments that cannot be trusted and a leadership team that stops believing the data.

Glossary

Terms this stage depends on.

Statistical power

The probability that an experiment will detect an effect of a given size if one truly exists. Standard practice: 80% or higher.

Minimum detectable effect (MDE)

The smallest true lift the experiment is powered to detect at the chosen significance level and power.

Unit of randomization

The entity assigned to variant (user, session, account). Chosen so that spillover between variants is prevented.

Gut-driven vs fully instrumented

What changes when this stage is done properly.

Illustrative comparison of experiment discipline.

MetricGut-driven attributionFully-instrumented data pipeline
Sample size"Let it run for two weeks."Calculated from MDE, baseline rate, power, and traffic.
RandomizationCookie-based, resets on logout.User-based, persistent across sessions and devices.
Analysis planWritten after the readout.Pre-registered before launch.
What we actually do

The operational shape of this stage.

  1. 01
    Compute required sample size against the pre-registered MDE, baseline, and power.
  2. 02
    Select the unit of randomization that avoids cross-variant contamination on the surface being changed.
  3. 03
    Verify primary and guardrail metric instrumentation with a shadow test before opening allocation.
  4. 04
    Pre-register the analysis plan, including stopping rules and interim look protocol.
Worked example
Illustrative

Sizing an A/B test on a 10,000-lead-per-month funnel

Baseline conversion 8%. Minimum detectable effect 20% relative (8% to 9.6%). At 80% power and a 5% two-sided significance level, the test requires roughly 8,500 users per variant. On 10,000 leads per month split 50/50, that is a roughly six-week cycle. Any test scoped for a lift below 20% relative on this surface will not read out in a reasonable window and should be re-scoped or replaced.

Questions

Frequently asked about this stage.

How do you size an A/B test?

+
Sample size is computed from the baseline conversion rate, the minimum detectable effect the business considers operationally meaningful, the desired statistical power (typically 80% or higher), and the chosen significance level (typically 5% two-sided). Traffic determines cycle length. Cycle length determines throughput.

What is a minimum detectable effect and how do I choose one?

+
The minimum detectable effect is the smallest true lift the experiment is powered to detect. Choose it based on what change in the primary metric would be worth acting on operationally, not on what would be nice to see.

See this stage run against your numbers.

A 30-minute Growth Audit. You leave with two or three specific findings, whether or not we ever work together.

Newsletter

Field notes on measurement, experimentation, and growth for health and SaaS. No fluff.

Occasional dispatches from real engagements — attribution, revenue engineering, digital-health growth. Sent when there is something worth reading, not on a drip schedule.

No spam, ever. One-click unsubscribe. We never share your email.
Book a Growth AuditFree calculator