Run A/B tests and personalize your store without code. 14-day free trial, no credit card.
Start free trial →In-depth guides and case studies where this concept is put to work.
Lift is the percentage improvement a variant achieves over the control on a specific metric in an A/B test. It answers the question: "By how much did the change improve (or worsen) performance?" Lift is the primary language of experimentation results — a 7% lift in conversion rate means the variant converted 7% more visitors, relative to the control's baseline rate.
Lift (%) = ((Variant Metric − Control Metric) / Control Metric) × 100
Example:
Lift can be calculated on any metric: conversion rate, average order value, revenue per visitor, add-to-cart rate, or checkout completion rate.
Lift converts statistical results into business language. A 10% lift in conversion rate on a Shopify store generating ₹30 lakh per month means ₹3 lakh in additional monthly revenue — without spending an extra rupee on ads. Tracking lift across experiments also helps D2C brands build a portfolio view of their optimization program: if your average experiment delivers 4% lift on conversion rate and you run 24 experiments per year, you can project the compounding revenue impact on your growth roadmap. Lift also surfaces which types of changes — copy, layout, pricing display — consistently move the needle for your audience.
The Man Company, a men's grooming D2C brand, tested adding a "Most Popular" badge to their best-selling face wash on the product listing page. The control had an add-to-cart rate of 6.1%. The variant with the badge reached 6.9%. Lift = ((6.9 − 6.1) / 6.1) × 100 = 13.1% lift. At their traffic volume, that translated to approximately ₹4.2 lakh in additional monthly revenue attributed to a single badge.
Lift is the output metric most commonly used to communicate A/B test results to stakeholders. In testing platforms, lift is usually displayed automatically alongside confidence intervals, which express the range within which the true lift likely falls. A lift of 8% with a confidence interval of [3%, 13%] means the true improvement is likely between 3% and 13%.
Run smarter A/B tests with CustomFit.ai — 14-day free trial, no credit card required.