Split Testing in TikTok Ads Manager

Measurement
Product and feature updates
|

Last updated: May 2025

|

10 minute read

New in Split Testing: Campaign-Level Split Testing in TikTok Ads Manager

We are excited to announce the launch of the Beta stage for Campaign-level Split Testing in TikTok Ads Manager, which now includes Smart+ and Custom-Variable testing! Along with UI improvements, you can expect an enhanced experience from the current ad group split testing setup experience.


What is Split Testing?

Split Testing in TikTok Ads allows you to A/B test two different ad groups or campaigns to determine which performs better. This method ensures accurate performance comparisons to help you make informed decisions and improve your campaigns. You can test five variables: Targeting, Bidding & Optimization, Creative, Smart+, and Multi-Variable.

Why use Split Testing?

Have you ever had two competing ideas for an ad campaign and wished you could try them both at once? Now you can do so with Split Testing! With campaign-level Split Testing, you can:

  • Compare different ad variations

  • Identify the most effective campaign settings

  • Refine your strategies and optimize future campaigns


Key Features

  • Compare Two Campaigns: Test different campaign structures, creative strategies, and objectives.

  • Audience Segmentation: The system randomly divides the audience into two equal groups (A & B) to ensure fair testing.

  • Exclusive Exposure: Each audience group sees only one version of the ad, preventing cross-exposure and maintaining data integrity.

  • Statistical Analysis: Metrics from both campaigns are measured, and the system determines the winning ad group based on 90% statistical significance.

  • Automated Smart+ campaign creation: The system automatically generates a Smart+ campaign based on the advertiser’s manual campaign for an accurate comparison.

  • Cross-Objective Comparisons: Test campaigns with different objectives (e.g., Reach vs. Traffic) to determine the most effective approach.


How It Works

TikTok Ads Manager randomly divides the audience pool into two equal-sized groups, A and B. Each campaign or ad group is shown only to its corresponding audience group: audience Group A will see only Campaign A, and audience Group B will only see Campaign B. Performance metrics are measured and compared between the two campaigns and the system determines the winning campaign using statistical significance.

Call to Action

Are you ready to try Campaign-Level Split Testing? You can get started in TikTok Ads Manager today!