What is A/B testing?
A/B testing (split testing) is a method of comparing two versions of a webpage or feature to see which one performs better. Visitors are randomly shown version A or version B, and data is collected to measure which drives more engagement, clicks, or conversions.
Example: Testing whether adding “Add to Cart” and “Choose Size” buttons increases purchases compared to a version without them.
The assumption you make before launching is your hypothesis. The test results will prove or disprove it—and tell you which variation wins.
Client-Side A/B Testing
Client-side A/B testing runs in the user’s browser using JavaScript. It’s perfect for testing visual or structural changes—like headlines, CTAs, images, button colors, or page layouts.
When someone visits your site, JavaScript modifies the page in real-time to show different variations. It’s fast, flexible, and doesn’t require backend development.
Benefits:
- Easy to set up—no coding required
- Perfect for marketers running quick conversion tests
- Fast iteration on front-end elements
Limitations:
- Flickering (FOOC): The original page may flash before the test variation loads, hurting user experience
- Cookie-dependent: Relies on browser cookies to identify users—challenging in a cookie-less world
- Performance impact: JavaScript snippets can slow page load times
- Surface-level only: Can’t test backend logic, algorithms, or features
Best for:
- Testing design, copy, and layout changes on websites.
Server-Side A/B Testing
Server-side testing happens on the web server before the page is sent to the user’s browser. It lets you test backend functionality, algorithms, and features—not just what users see.
Benefits:
- No flickering: Variations are pre-rendered, so users see a seamless experience
- Faster performance: Minimal impact on page load time
- Deep testing: Test algorithms, pricing models, search logic, payment methods, and more
- Omnichannel: Run experiments across web, mobile apps, email, and connected devices
- Greater control: Full access to server-side code and user tracking (no cookie dependency)
- Secure: Sensitive data stays on the server
Use Cases:
- Mobile app testing: Test features in native apps without relying on browsers
- Algorithm optimization: Compare recommendation engines or search algorithms
- Pricing experiments: Test shipping costs, subscription models, or payment options
- Feature rollouts: Use feature flags to release new features to a subset of users first
Best for:
- Testing product functionality, backend logic, and cross-platform experiences.
Check out these 50 Tests for Website Optimization! →
Client-Side vs Server-Side: Which Should You Use?
| Factor | Client-Side | Server-Side |
|---|---|---|
| Setup | Easy, no coding needed | Requires technical setup |
| Best for | Visual/UI changes | Backend features & logic |
| Performance | Can slow page load | Minimal impact |
| Flickering | Possible | None |
| Tracking | Cookie-based | User ID or session-based |
| Scope | Websites only | Web, mobile, email, IoT |
| Control | Limited to front-end | Full stack |
The answer? It depends on your goals. Many teams use both together for maximum impact—client-side for quick UI wins, server-side for deep product experimentation.
Learn more about server and client-side a/b testing here →
A/B Testing with Feature Flags
Testing Beyond the Surface
A/B tests are good enough to test “cosmetic” changes to your website, but what happens when you want to test out backend functionality and new features? Enter A/B tests driven by feature flags.
This means you can test out your functional features and then target users by dividing them into various segments. You’ll define variations and assign percentages to each of the variations. In this case, variations are made using a combination of different flag values.
You can use all available flags in your system and combine them to test out these various combinations and how your users respond to them. By delivering different, simultaneous versions of your feature to predetermined portions of traffic, you can observe and use the data extracted to determine which version is better.
Experiment Toggles
Feature flags or toggles used in this way are also called experiment toggles. These are highly dynamic and are likely to remain in the code a few days or weeks—long enough to generate statistically significant results required for A/B tests.
Using Feature Flags to Conduct Server-Side Testing
Feature flags are a software development tool that allows you to decouple deployment from feature release—making releases quicker and less risky.
With feature flags, you can conduct server-side A/B tests by rolling out new features to a small group of users. You can then measure their performance on this group of users before rolling out to everyone else.
If you want to try your hand at server-side testing, AB Tasty offers server-side experimentation in order to conduct safe experiments for your features by setting up user-targeting and KPIs to track and put you on the road to quicker and safer releases.
Ready to go further? Let’s build better experiences together →
