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.
The treatment group (also called the variant, Variant B, or the challenger) is the segment of visitors in an A/B test who experience the modified version of a page, element, or experience. This is the group that sees your hypothesis in action — the new CTA button color, the redesigned product page, the updated pricing layout. By comparing the treatment group's behavior against the control group, you can measure whether your change had a positive, negative, or neutral effect on your target metric.
The lift generated by the treatment group is calculated as:
Lift = ((Treatment CR − Control CR) / Control CR) × 100
Where CR = conversion rate. For example, if the control group converts at 3.0% and the treatment group converts at 3.6%:
Lift = ((3.6 − 3.0) / 3.0) × 100 = +20% relative improvement
This lift becomes actionable only when it crosses the threshold of statistical significance.
The treatment group is where your CRO hypothesis meets reality. Every change you test — whether it's a new trust badge, a different image, or a simplified checkout form — gets its first real-world validation through the treatment group. For D2C brands, the treatment group gives you evidence-based answers to questions you'd otherwise argue about in product meetings: Does urgency messaging actually increase purchases, or does it annoy customers? Does showing EMI options increase AOV? Does an exit-intent popup hurt or help conversion? The treatment group data answers these questions with your actual customers, not assumptions.
Bellavita ran an A/B test where the treatment group saw a product page with a "Try Before You Buy" messaging banner (highlighting their return policy) added above the fold. The control group saw the standard page without this banner. After 12 days and 9,400 sessions per group, the treatment group showed a statistically significant 14% improvement in add-to-cart rate. The treatment group result was specific: the banner improved purchases for new visitors but had no effect on returning customers — a segmentation insight that shaped how the winning change was implemented.
The treatment group is the engine of your experimentation program — it's where learning happens. A well-designed treatment tests a specific, hypothesis-driven change, runs against a clean control group, and generates insights that inform future tests regardless of whether it wins or loses. A losing treatment is still a valuable signal: it tells you what your customers don't respond to, which is equally important for building a better store.
Run smarter A/B tests with CustomFit.ai — 14-day free trial, no credit card required.