📘 Support Doc

A/B Split Testing PRO

Run two squeeze pages side-by-side and let real traffic tell you which one collects more opt-ins. ConvertPages handles the traffic split, conversion tracking, and confidence math automatically.

What you can test

A/B testing in ConvertPages is for squeeze pages only. The metric being measured is opt-ins — how many visitors actually submit their email. We chose this scope because it's the only metric that genuinely matters for a squeeze page, and the only one that's directly measurable on your end.

💡

Bridge pages are intentionally not supported. Their value is sending traffic to an offer page elsewhere — and we can't track conversions that happen on someone else's domain. So a "click-through rate" test would be measuring activity, not results.

How it works

You publish two squeeze pages — Variant A (your control) and Variant B (your challenger). They can differ in headline, button colour, image, copy — anything you want to test.
You create an A/B test in ConvertPages, pairing the two pages.
You're given a single test URL. Send all your traffic to that URL.
ConvertPages randomly serves Variant A or B to each visitor (50/50 by default), and remembers which one each person saw.
Opt-ins are recorded against whichever variant the visitor saw.
The dashboard shows live stats: visits per variant, opt-ins per variant, and conversion rate for each.
Once you have enough data, ConvertPages tells you whether one variant is statistically significantly better — and you can declare a winner.

Setting up a test

1. Build both pages first

Before opening the test creation modal, make sure both squeeze pages are published. The easiest workflow:

Build and publish your control page (Variant A)
From My Pages, duplicate it using the duplicate button
Open the duplicate, change the one thing you want to test (e.g. headline)
Publish the duplicate

Now you have two near-identical squeeze pages with one controlled difference between them.

2. Create the test

Open the editor and click the A/B Tests button
Click + Create A/B Test
Give your test a memorable name (e.g. "Headline test — emoji vs no emoji")
Pick Variant A from the dropdown — only your published squeeze pages appear here
Pick Variant B from the dropdown
Leave the traffic split at 50/50 unless you have a specific reason to skew it
Click Create Test

You'll get a public test URL — that's the one you share with traffic. Visitors hitting that URL get rotated between A and B.

Reading the stats

Each variant card shows:

Conversion rate — opt-ins ÷ visits, expressed as a percentage
Opt-ins / visits — the raw counts behind the rate

Below the cards, a confidence indicator tells you how much weight to put on the result:

Low confidence — keep collecting data. We need at least 30 visits per variant before the math is meaningful.
Medium confidence — there's a likely winner but it's not certain yet.
High confidence — the difference is statistically significant. Time to declare a winner.

⚠️

Don't declare a winner too early. Small sample sizes can show big false differences. Wait for at least medium confidence — ideally high — before stopping the test.

Declaring a winner

When you're confident in the result, click Declare winner on the winning variant. This:

Stops the rotation — 100% of future traffic goes to the winning page
Locks the test (you can no longer change the variants)
Keeps the historical stats visible for reference

The test URL keeps working — it just always serves the winner now. So your existing campaigns/links don't need to change.

Tips for useful tests

Test one thing at a time. If you change the headline AND the button colour AND the image, you won't know what caused the difference.
Test things that matter. A button colour change rarely moves the needle. Headline, hero image, and the offer itself usually do.
Give it time. A few hours of low traffic isn't enough. Aim for at least a few hundred visitors per variant before drawing conclusions.
Don't run forever. Once you hit high confidence, declare a winner. Letting a test run indefinitely just wastes traffic.

Troubleshooting

Both variants stuck at 0% conversion rate

Make sure both pages are squeeze pages (not bridge pages) and that the form on each page is connected to a working integration (Aweber, GetResponse, etc). The conversion tracker counts form submissions — if no one's submitting, there's nothing to count.

Stats aren't updating

Stats refresh when you reload the A/B Tests panel. Refresh the page to see the latest numbers.

I want to test a bridge page

Not supported. See the What you can test section above for why.

Need help?

If something's not working as expected, contact support.