Ship the Version That Actually Wins
Set up an experiment against a goal you already track, split traffic across variants, and read a clear result. No second analytics tool, no waiting for a free engineer.
From idea to result in four steps
Experiments live in your dashboard under Domains > your site > Experiments.
Pick a success metric
Choose one of the goals you already track as the metric the experiment is judged on. No new instrumentation and no second definition of what counts as a conversion.
Add your variants
Create two or more variants and set the traffic split between them. One variant is the control that every other variant is compared against.
Launch when it looks right
An experiment is saved as a draft first, so you can review the split and the metric before a single visitor is assigned. Launch when you are ready.
Read the result
Watch per-variant conversion rates, the confidence interval on lift over control, and how the trend moves over the days the test runs.
What you actually get
Concrete limits and outputs rather than promised results.
Experiment capabilities
What the experiment engine does for you, and the two things it asks of you.
Deterministic Variant Assignment
Each visitor is assigned to a variant by the tracker, using a hash of their visitor ID and the experiment. The same visitor keeps the same variant on every page and on every return, so nobody watches the layout flip mid-journey. Assignment is computed in the browser, so there is no extra network round trip and no flicker while a server decides. A/B/n is supported: add as many variants as the question needs and give each one its own share of traffic.
Tied to Goals You Already Track
The success metric is an existing goal, so an experiment reads the same conversion data as the rest of your dashboard. Two tools can no longer disagree about the number.
Frequentist Statistics
Per-variant conversion rates with a 95% confidence interval on lift over control, so you can tell when a difference is still inside the noise.
Results Trend Over Time
See how each variant performed day by day instead of only as one cumulative number, so a result that came from a single unusual week does not go unnoticed.
One Line of JavaScript
Zenovay decides which variant a visitor is in; your site applies it. That means a small piece of JavaScript to swap the headline, the button or the layout. There is no visual editor.
Which plans include A/B testing?
A/B testing is available on Pro and above, with up to 5 concurrent experiments on the Pro plan. It is not included in the Free plan.
Do I need a developer to run an experiment?
You can create, launch and read an experiment entirely from the dashboard. Applying the variant to the page does need one line of JavaScript on your site: Zenovay tells you which variant a visitor is in, but it does not rewrite your page for you.
How does it behave in cookieless mode?
In cookieless mode there is no persistent visitor identifier, so variant assignment is scoped to the browser tab and resets when the tab is closed. The same person can see a different variant in a later session. If assignment has to persist across sessions, run the experiment on a site where cookie-based identification is enabled.
Stop Guessing Which Version Is Better
Run your first experiment against a goal you already track.
Pro and above • 5 concurrent experiments on Pro • Cancel anytime