Framer improved its built-in analytics with custom distribution for A/B tests. Instead of splitting traffic evenly, you can now send a variant to a smaller slice of visitors first, watch how it performs, and scale up only when the numbers support it. It is a practical upgrade for anyone who treats their marketing site as something to optimize, not just publish.

What custom distribution does

A standard A/B test often splits traffic evenly between versions. Custom distribution lets you decide the split yourself, so you can expose a new variant to a smaller audience before committing to a wider rollout. That makes it easy to trial a redesign, a new hero, or a different call to action on a controlled portion of visitors, measure the result, and expand with confidence. It arrived alongside a handful of other analytics improvements in the same update.

Why it matters for real projects

Rolling a big change out to everyone at once is a gamble. If the new version underperforms, you have already exposed your whole audience to it. A smaller initial distribution acts like a safety valve: you get real data from live traffic while limiting the downside. Because the analytics are built into Framer, you can run these tests without stitching together external tools, which keeps the loop between idea, test, and decision tight.

Who it helps

• Marketers optimizing landing pages and conversion paths.

• Teams testing a redesign before a full launch.

• Founders who want evidence before betting the whole site on a change.

How to use it

Set up your A/B test in Framer, then assign a custom distribution so the variant starts with a smaller share of traffic. Let it run long enough to gather meaningful data, review the results in analytics, and increase the variant's share as your confidence grows. Treat each test as a small, reversible step toward a better performing site.

Our team pairs design and development with growth, so testing is part of how we ship rather than an afterthought. If you want a Framer site that improves through measured experiments instead of guesswork, Book a discovery call and we will build a testing plan around your goals.