A campaign can hold more than one version of the popup. Each version is a variant. Once two or more of them are switched on, your traffic is split between them. That is the A/B test.

Where variants live

Open a campaign and stay on the Overview tab. Under the metrics you will find the variant table, with one row per variant.

The desktop table has these columns: Traffic, Variants, Active, three performance columns that follow the campaign's goal, and Uplift. With the default Submit goal, the performance columns are Impressions, Submits and Submit rate. With the Order goal, they are Visitors, Orders and Order rate.

Each row shows a thumbnail of the variant, its name, its traffic share as a percentage, and a switch that turns it on or off. While a variant is still being built, the row says Building… instead of showing a preview.

Adding a variant

Under the table there is Add A/B test variant. It opens a menu:

  • Control (no popup): "A holdout bucket that sees nothing". These visitors get no popup at all, so you can measure the popup against showing nothing. You can only have one. If you already have it, the menu says Already added.

  • Duplicate a variant: "Clone an existing design to tweak". Pick one of your designs, and you get a copy named after it, ready to edit.

  • Use an earlier concept: finished mockups from this design session that were never built. This option only shows up when there is at least one.

  • Write a variant: "Generate a new design in the designer".

There are two more routes:

  • Each row's own menu has Rename, Duplicate, New design from this variant and Delete.

  • On a design session's concept board, you can select finished concepts and use Add {count} as A/B variants to attach them to an existing campaign.

A concept added from the campaign chat arrives switched off: "Variant added switched off. Turn it on in the A/B table when ready." Switch it on in the Active column when you want it to take traffic.

Splitting the traffic

By default, traffic is shared equally across all active variants.

To change that, click the percentage in the Traffic column. This opens A/B test settings. Under Traffic share, type a percentage for each active variant and click Save. Each variant needs 1–99%, and the total must be 100%. To go back to an equal split, tick Evenly split.

If you switch a variant off and back on later, it gets its old share back.

In two cases the split can't be edited, and the panel tells you why:

  • the campaign's OptiMonk connection doesn't support it yet

  • the campaign is part of an experiment

Asking the assistant for test ideas

The chat next to the campaign can propose variants. Ask something like "Suggest an A/B test I could run". It is one of the suggested prompts. The assistant may first ask a couple of questions. Then it comes back with a set of labelled ideas, each with a preview.

Click an idea to open Variant comparison. Your Control · current is on one side, the Suggested version on the other, plus Why this works and What changes. Add variant creates it in the campaign, and the button then reads Added. If none of them appeal, there is None of these — show me other variants.

Editing a variant

Click a variant row to open it. The panel on the right has two tabs: AI chat, where you describe a change in words, and Editor, where you edit by hand.

Nothing here reaches your visitors on its own. Every change is saved as a draft: "Nothing changes for your visitors until you press Publish." Publish is the button that takes the change live.

If two variants still share one build, opening the editor makes a copy first. That way, editing one variant never changes the other.

Running the test

There is no separate "start test" button. Once two or more variants are active on a live campaign, a banner appears above the table saying A/B test running. Until there is enough data, it reads Data collection in progress. After that it names the leader, for example "Variant B is leading by +12.40% (87% chance to win)".

The leader is always one of your designs. The Control (no popup) is the yardstick, so it can't be declared the winner. If your designs don't beat the control, pause the campaign instead.

Ending the test by hand

The button on the banner ends the test. It keeps the leading variant serving and switches off all the others, including the control.

  • Without automatic declaration, the button reads Declare winner at {threshold}%.

  • With automatic declaration switched on, it reads Declare now.

If the leader's chance to win is below your threshold, or there is no data yet, you get a confirmation first: Declare winner now? with Keep testing and Declare anyway.

Letting OptiMonk end it

In A/B test settings, switch on Auto-declare winner: "Automatically ends the test when a clear winner emerges, and stops a clear loser." It is off by default.

You can also switch it on from the banner. If the leader is still below the threshold, the banner offers Watch at {threshold}%, which turns it on.

You set three conditions:

  • Declare at … chance to win (default 90%, between 50% and 99%)

  • Min conversions (default 100)

  • Min days running (default 3)

OptiMonk checks the test once a day, using the full days since your last change to the test. Two things can happen:

  • If the leading design has reached your chance-to-win threshold, has at least the minimum conversions, and the test has run for at least the minimum days, it is declared the winner. It becomes the only active variant.

  • If, under the same minimums, a design's chance to win drops very low (below 100% minus your threshold, so below 10% at the default 90%), that design is switched off. The test keeps running with the rest.

Some things to know:

  • The control is never switched off or declared by the automatic check. With fewer than two designs, it does nothing.

  • Automatic stopping only works when the campaign goal is Submit. For other goals the panel says so, and the switch can't be turned on.

  • While it is on, the banner shows Watching {threshold}% and the time of the last check.

  • When OptiMonk stops or declares a variant, a notice on the campaign tells you which one, with its chance to win and conversions.

Where the results show up

The variant table is the scoreboard. It shows the performance columns and Uplift per variant, with the chance to win next to the leader. Above it, the metrics band shows the campaign totals for the goal and date range you pick. The campaign also has an Analytics tab.

Some things to know about the numbers:

  • The date range. The table follows the date range you pick above it. While a test is running, the default range is the current test phase. Any change to the test opens a new phase from the next day: switching a variant on or off, changing the split, or declaring a winner. The trend chart marks it with "New A/B phase · change: {date}". This keeps visitors who saw a different mix of variants out of the comparison.

  • The baseline for uplift. Uplift is measured against the control if you have one. Otherwise it is measured against the weakest-converting variant that has traffic, so the uplift shows how far ahead the others are.

  • Chance to win is a Bayesian estimate. It needs a reasonable amount of data before it means much.