New research guide: The Confidence Economy, how AI is changing fashion commerce.
Read it →
Measurement
A/B Testing
A/B testing is a controlled experiment that splits traffic between two variants — A (control) and B (test) — to measure which produces better outcomes, using statistical significance to confirm that results are not random.
What this means
A properly run A/B test randomly assigns traffic to each variant, runs long enough to reach statistical significance, and isolates a single change between variants so the cause of any difference in outcomes is clear. Poorly run tests — small sample sizes, multiple simultaneous changes, non-random assignment — produce results that look conclusive but aren't. For AI shopping tools specifically, the relevant A/B test compares shoppers with access to the tool (B) against a genuine random holdout without access (A) — not a comparison between shoppers who voluntarily used the tool and those who didn't, which introduces selection bias.
Why it matters for fashion brands
A/B testing is the standard methodology for measuring conversion improvement on ecommerce stores. For AI tools, a properly designed A/B test (with a genuine holdout control group, not a voluntary-use comparison) is the only way to know whether an AI stylist is producing incremental revenue or just correlating with high-intent sessions.
Related terms
Holdout-Based Lift Report — A measurement method that isolates a tool's true incremental impact using a randomized control group. Incremental Lift — The actual revenue improvement caused by a tool, isolated through a holdout test. Conversion Rate — The percentage of website sessions that result in a purchase.
See how Elara measures results
See how Elara's holdout-based lift report applies rigorous testing methodology.