How to Run A/B Split Testing in Framer with Humblytics

A/B - or split-testing lets you compare two designs under real traffic to see which one drives more conversions. By pairing Framer’s rapid-prototyping workflow with Humblytics’ privacy-first analytics, you can:

• Boost conversion rates without guesswork

• Base decisions on statistically significant data

• Understand user behavior at a granular level

Why A/B Testing Matters

Before we jump in, let's quickly touch on why A/B testing is so crucial:

• Boost conversion rates

• Make data-driven decisions

• Understand your audience better

• Maximize your ROI

Not set up with analytics yet? Start here to add Humblytics to your Framer site.

1. Setting Up Your Framer Project

1. Duplicate your original page inside Framer.

2. Name one Variant A (control) and the other Variant B (challenger).

3. Publish the project; Framer assigns a unique URL to each page (e.g.,
   https://your-site.framer.website/landing-a and …/landing-b).

4. Copy these URLs—you’ll need them when you set up the experiment in Humblytics

Prepare Your Elements

• Clicks

• Sign-ups

• Any other conversion metrics

Publish Your Project

Hit that publish button and grab those unique URLs for each variant. You'll need these for the A/B test setup.

2. Integrating Humblytics with Framer

1. Sign Up or Log In to Humblytics:

   • If you don’t have a Humblytics account yet, sign up at Humblytics. If you already have an account, simply log in.

2. Add your Framer Site: Follow the Framer Analytics Guide for detailed instructions on integrating Humblytics with your Framer site.

3. Track Custom Events: Learn how to set up custom event tracking by visiting the General Guides Page on Humblytics.

3. Setting Up the A/B Test in Humblytics

1. Navigate to A/B Testing: In your Humblytics project dashboard, find the Split Testing section. Click on "Start New Experiment."

2. Configure Your Test: Give your test a descriptive name and enter the URLs of your Variant A and Variant B pages from Framer.

3. Set Goals: Define the goals for your test, such as click-through rates, form submissions, or any custom events. Specify the key performance indicators (KPIs) to measure success.

4. Select Randomization: Choose between session-level or user-level randomization based on your needs.

5. Launch the Test: Review your test settings and click "Start Experiment" to begin the split testing process.

4. Analyzing Split Test Data

Real-Time Reporting

• Visitors, page-views, unique users

• Conversion rate per variant

• Bounce rate and average scroll depth

Ensuring Statistical Significance

Humblytics displays a live confidence score. Industry standard is 95 %. If your lift is +12 % but confidence is 70 %, keep the experiment running until either:

• The confidence passes 95 % or

• You hit the pre-calculated sample size (use Humblytics’ built-in sample-size calculator).

1. Access Experiment Results

   • Log in to Humblytics and head to the "Split Testing" section.

     
     

2. Overview Section: Get a bird's-eye view of your experiment:

   • Unique visitors

   • Page views

   • Conversion rates

   • Bounce rate

   • Average session length

   • Average scroll depth
     
     

3. Target Clicks Section

   • See how those buttons and links are performing across variants.
     
     
     

     

     
     
     

4. Form Submissions Section

   • Compare form submission data between your variants.
     
     

5. Real-Time Monitoring

   • Watch those conversions roll in as they happen!
     
     

6. ‍Determining the Winner

   • Analyze the data against your goals and KPIs.

   • Make sure your results are statistically significant before popping the champagne.
     
     

7. Implementing the Best Variant

   • Once you've crowned a winner, implement it on your Framer site.

   • Keep monitoring to ensure continued success.

Step 5 - Analyze Results & Declare a Winner

1. Compare lift: control vs variant for each KPI.

2. Segment by device, traffic source, or geography to spot hidden patterns.

3. If the variant wins with ≥ 95 % confidence, click Promote Variant—Humblytics
   automatically routes 100 % of traffic to the winner.

4. Multivariate testing: if you have more than one element to tweak (e.g., headline and image), spin up a multivariate experiment to test combinations without creating dozens of manual variants.

Continuous Optimization Roadmap

• Monthly growth sprint: pick the next bottleneck (e.g., pricing page) and launch a new test.

• Document learnings: keep a shared spreadsheet or Notion page of hypotheses, results, and insights.

• Heat-maps + session replays (available on Growth plan) reveal qualitative clues for your next experiment.

• Re-test winning variants annually; user behavior and traffic sources evolve. How long should an A/B test run?

Until you achieve 95 % statistical confidence and meet your minimum sample size. Humblytics shows both metrics in real time.

What if both variants perform the same?

If the lift is < 1 % and confidence is high, keep the control. Test a more substantial change next.

Can I segment results by device or referrer?

Absolutely. Use the Segments dropdown inside the experiment report.

Does Humblytics slow my Framer site?

No. The 36 KB script is served via a global CDN and loads asynchronously after the main thread.

Conclusion

You're now equipped to run A/B split tests like a pro on your Framer website using Humblytics. By following this guide, you'll be making data-driven decisions that can significantly impact your website's performance and user engagement.

Remember, the key to success is continuous iteration and optimization. So go forth, test boldly, and may your conversion rates always be in your favor!

Have questions or want to share your A/B testing wins? Drop us a line at support@humblytics.com. We'd love to hear from you!

Next Steps

You’re now equipped to launch statistically sound, privacy-first A/B tests in Framer. Start a free 7-day Humblytics trial, run your first experiment, and watch conversion rates climb—no cookies, no extra plugins, just data-driven growth.

Ready to optimise? → Start your trial now.