How to Create a Widget Test
Nelio A/B Testing is the only A/B testing tool that lets you test WordPress widgets. You can easily duplicate your current widgets, modify them, and measure which combination delivers the best results.
Widget tests help you discover which widgets work best on your website — for example, whether a Featured Posts widget or an Ad Banner performs better. They’re also ideal for optimizing ads, since most ad placements in WordPress appear as widgets.
Table of Contents
Step 1: Create a New Widget Test
- In your WordPress dashboard, go to Nelio A/B Testing → Tests.
- Click Add Test.
- In the dialog, select Widget (under the “Global” test category).

💡 This will create a test based on the set of widgets currently active on your site.
Step 2: Name and Define Your Test
Give your test a clear, descriptive name (for example, “Sidebar Widgets – Ad vs. Categories”).
In the right sidebar you’ll find:
- Status & Management — Shows whether the test is Draft, Ready, Scheduled, or Paused, and lets you set start/end dates.
- Description — Briefly describe what you are testing and what you expect (e.g., “We believe the new theme layout will increase product views”).
- Test Scope — Specify where the widget variants will appear. By default, variants are visible across your entire site. To limit the scope to certain pages, use the URL form to define where the test should run (for example,
/shop/*, or/blog/).

Step 3: Create and Edit the Variants to Test
A widget test measures which widget configuration performs best for your defined goals. For instance, you might test whether showing featured posts in the sidebar instead of ads results in more clicks or conversions.
Create New Variants
- Create a “Variant B” (or more than one) of your active widget setup.
- Each new variant starts as an exact copy of your current widgets.
- Rename each variant for clarity — for example, “Sidebar with Ads” or “Variant 2 – Newsletter Block.”
- Click Edit to customize each variant. You can also click Preview, Duplicate, and Reset.

💡 Click New Variant to create a new alternative based on your current widget set, or use Duplicate under any existing variant to create a copy of that specific configuration.
Editing a Variant
Editing widget variants is simple — Nelio uses WordPress’s own widget editor, with a few additional controls:
- Add, remove, or rearrange widgets as you normally would and click on Update when you’re done.
- Use the Duplicate Control Widgets button (at the top of the editor) to overwrite a variant with the control widgets if you want to start fresh.

Step 4: Define Goals and Conversion Actions
Every test needs a goal — what counts as success.
Each goal can include one or more conversion actions that describe the specific behavior you want to track.
Name your goals, then click the Add Action button to choose from the available conversion actions.

You can combine multiple actions under one goal.
| Category | Example Actions |
|---|---|
| View | Page view, Element view |
| Interaction | Click, Click external link, YouTube Video Playback, Scroll, Engagement |
| E-Commerce | Add WooCommerce product to cart, WooCommerce Purchase |
| Other | Time on Page, Custom event |
🔍 Learn more: What Are the Conversion Actions and Goals? →
Step 5: Add Segmentation Rules (Optional)
Segmentation lets you decide who participates in the test. You can target visitors by conditions such as:
URL parameter · Referrer · Language · Location · Logged-in status · Visitor Type · Day/time · Browser · Device type · Operating system · Window Width · Cookie · IP address
If you skip segmentation, all visitors to the tested pages are included automatically.
🔍 Learn more What is Segmentation and How to Use It →
Step 6: Start the Test
When everything is ready:
- Review your setup summary at the top of the page.
- Click Start.
Nelio A/B Testing will:
- Randomly assign visitors to variants.
- Track conversions anonymously via Nelio Cloud.
- Display real-time results in your dashboard.
⏯ Pause, Resume, or Stop Your Test
- Pause a running test if you need to make changes, then resume it later.
- Stop a test manually whenever you want, or let it end automatically under certain conditions such as duration, total number of visits or confidence results (available in Professional and Enterprise plans).
- Once a test is stopped, its results remain available for review.
⚠️ Important note about editing paused tests
Editing a paused test may reset or affect ongoing statistics. When in doubt, duplicate the test before resuming.
Step 7: View Results and Analyze Performance
While your test runs, go to Nelio A/B Testing → Tests and click View Results to monitor its progress.
Summary of the results
In the right sidebar, you’ll find a quick overview of your test:
Goal selector – Choose which goal’s results to display.
Status icon – Indicates experiment status:
🟡 Collecting data.
🕒💛 Potential winner, low confidence
🟢🏅 Clear winner found.
🔴 No improvement over control.
Unique results – Filter duplicates.
Public results – Share a public link.
Export results – Download a CSV.
Summary – Shows status, duration, start date, and end condition.

Test Progress

The main area of the results page is divided into two tabs that visualize your test’s progress and performance:
- Conversion Rates — A line chart showing conversion evolution across variants.
- Views and Conversions — Compares total visits and conversions for each variant.
Confidence indicators appear alongside these charts, updating automatically as more data accumulates.
Below the charts, you’ll find a summary table with:
- A thumbnail preview (if available) of each variant.
- Number of visits and conversions.
- Conversion rate and improvement over control.

Step 8: Improve Your Web Conversion
You can end a test manually using the Stop button in the upper-right corner of the results page, or let it end automatically based on the configured end mode.
After completing a test and to improve the conversion of your website, click on the Apply button of the winning variant (if it exists) and the content of the original widget will be replaced with the contents of the winning variant.
Did you find this useful?
Thanks for your feedback!
Not what you’re looking for? Ask us directly:


