Flint logo
Back to guides
Tutorials

A/B Testing with VWO on Flint Pages

VWO is a visual experimentation platform that lets you create and run A/B tests on your Flint landing pages without writing code. This guide walks you through the full setup.

Overview

VWO is a conversion rate optimization (CRO) platform with a point-and-click visual editor. Once you install the VWO script on your Flint site, you can create test variants, define goals, and track results entirely from the VWO dashboard - no developer required for most tests.

The integration works at two levels:

  • •Component-level tests - Test different headlines, CTA copy, button colors, or any element within a page
  • •Full page variants - Test two completely different page designs against each other using VWO's Split URL testing

What you need

  • •A Flint account (sign up here)
  • •A VWO account (vwo.com) - VWO offers a free trial
  • •Your VWO Account ID, which you can find in your VWO dashboard under Settings > Account Settings

Step 1: Add the VWO script to your Flint site

VWO requires a single JavaScript snippet installed in the <head> of every page you want to test.

  1. 1.In your VWO dashboard, go to Settings > Implementation and copy your VWO SmartCode snippet. The SmartCode is unique to your account and includes your Account ID — copy it exactly as shown; do not edit it.
  2. 2.In your Flint editor, click the Settings button in the top navigation bar, then open Site Settings > Custom Code.
  3. 3.Paste the VWO SmartCode into the Head field.
  4. 4.Click Save. The script is now active on all your Flint pages.

Important: Always copy your SmartCode directly from VWO's dashboard rather than writing it by hand. The snippet is account-specific and contains settings that VWO manages for you.

Step 2: Create your first A/B test in VWO

Once the script is installed, go to VWO and create a new campaign:

  1. 1.In VWO, click Create and choose A/B Test.
  2. 2.Enter the URL of the Flint page you want to test (e.g., https://yourdomain.com/landing-page).
  3. 3.VWO will load a visual editor pointing at your live Flint page.
  4. 4.Use the visual editor to click on any element and edit it - change headlines, button text, images, colors, or layout. Each change creates a Variation.
  5. 5.Set your primary goal (e.g., button clicks, form submissions, page visits) by pointing VWO at the element or URL you want to track.
  6. 6.Configure your traffic split (e.g., 50% Control, 50% Variation A).
  7. 7.Set any audience targeting rules if needed (device type, traffic source, UTM parameters, etc.).
  8. 8.Click Start to launch the experiment.

Step 3 (optional): Run a full page variant test

If you want to test two completely different page designs, use VWO's Split URL Test instead of the standard A/B test:

  1. 1.In Flint, duplicate your landing page to create the variant (e.g., /landing-page-v2).
  2. 2.Make whatever changes you want on the variant page in Flint.
  3. 3.In VWO, create a new Split URL Test.
  4. 4.Set the control URL to your original page and the variation URL to your variant page.
  5. 5.VWO will route visitors between the two URLs and attribute conversions to each.

This approach is ideal for testing major redesigns or different messaging angles.

Step 4: Track conversions

VWO can track conversions automatically or through custom events:

Automatic tracking
- Page visit goals: VWO fires when a visitor lands on a specific URL (e.g., a thank-you page after a form submission).
- Click goals: VWO tracks clicks on any element you select in the visual editor.

Custom event tracking
If you want to track a specific interaction not covered by the visual editor, you can ask Flint to add a custom VWO event. VWO's current API uses the window.Wingify namespace (VWO is built on Wingify infrastructure). For example:

text
Add a VWO custom event to the CTA button on the /landing-page page.
When the button is clicked, fire:
  window.Wingify = window.Wingify || [];
  window.Wingify.event('cta_clicked', { page: 'landing-page' });
VWO (Wingify) is already loaded via the SmartCode script tag.

Paste that prompt into the Flint chat and the agent will instrument the button for you. The event name (e.g., cta_clicked) must match the custom event you've defined in VWO under the Data360 > Events section of your dashboard.

Reading your results

VWO uses Bayesian statistics (SmartStats) to determine a winner. Once your test has reached statistical significance:

  1. 1.Go to Reports in VWO to see conversion rates for each variation.
  2. 2.VWO will show you which variation is the winner and by how much.
  3. 3.If a variation wins, apply the winning changes directly to your Flint page by editing it in Flint (making the winning copy permanent), then stop the VWO test.
  4. 4.If neither variation wins after sufficient traffic, consider testing a more impactful change.

Tips for better results

  • •Test one meaningful change at a time (headline vs. headline, not headline + image + CTA simultaneously) - this makes it easier to understand what drove the result
  • •Run tests until you reach at least 95% statistical significance and a meaningful number of conversions (100+ per variation is a common minimum)
  • •Prioritize tests that address your biggest conversion bottleneck first - check your analytics to find the step where visitors drop off
  • •VWO's built-in heatmaps and session recordings can help you understand visitor behavior before you decide what to test