Let's meet at DMEXCO in Cologne!More
Teqblaze
Rate this article
Rating: 0 / Total: 0
Rating: 0 / Total: 0
Share this article
Let’s talk

We build AI-driven AdTech ecosystems for smarter monetization.

Homepage / Blog / Company news/
TeqBlaze enhances A/B Testing with more granular experimentation
Company news

TeqBlaze enhances A/B Testing with more granular experimentation

TeqBlaze enhances A/B Testing with more granular experimentation
August 25, 2026
4 min read
Let’s talk

We build AI-driven AdTech ecosystems for smarter monetization.

As SSP traffic volumes grow, even small changes in routing logic, filtering, or optimization models can affect sRPM, DSP spend, win rates, and other performance indicators. Platform owners need to validate those changes without applying them to the entire bidstream at once.

In version 2.5 of the TeqBlaze white-label SSP, the A/B Testing tool supported a basic global traffic split. The latest update expands this functionality with testing for specific supply–demand combinations, more precise traffic allocation, parallel scenarios, and improved reporting.

What has changed since version 2.5

The previous version of A/B Testing could identify performance changes but offered limited visibility into how specific inventory or demand paths contributed to the result.

Previously (v2.5)

Now

Comparisons were limited to global traffic splits.

Scenarios can target specific supply–demand combinations.

Experiments could not be scoped to individual supply–demand pairs.

When setting up a scenario, the system displays only the supply and demand sources supported by the selected optimization feature.

Parallel testing of different optimization features was limited.

Multiple scenarios can run in parallel, each with its own traffic allocation.

Reporting relied on the general Reporting module, with no scenario-level overview.

Up to five configurable preview metrics can be monitored directly in the A/B Testing section.

Scenarios were not preserved as standalone entities.

Scenario history is retained, including configurations and results.

The update also expands Range Margin testing. Demand and Supply are now available as separate options, allowing you to evaluate each side of the auction independently. Platform administrators can also configure the value the Main Traffic group uses for comparison directly within the scenario.

Extended CSV reports are available for further analysis and documentation.

CTA banner to the A/B testing tool pageRun controlled tests on specific parts of the stack

  • Built-in optimization functionality, including Traffic Shaping, Range Margin, SmartFloor, and VCR optimization, can be tested through dedicated scenarios. SmartFloor testing is also available for Direct Suppliers.

  • A scenario can be limited to specific inventory routed to a selected demand source or group of demand partners. The same supply–demand pair cannot participate in multiple active scenarios simultaneously, preventing overlap between experiments.

  • Separate scenarios can also test the same functionality with different traffic splits — for example, 10% and 40% — to compare performance under different levels of exposure.

  • Start and end times are configured for each scenario, allowing tests to launch and stop automatically within a defined period.

  • Manage launched, completed, and paused scenarios from a dedicated A/B Testing section. You can revisit, clone, repeat, stop, or delete previous scenarios, while their configurations and results remain available for comparison.

A/B Testing Tool settings screenThe interface of the updated A/B testing tool — screenshot from our white-label supply platform

Test changes before scaling them

Applying a new configuration to only part of the traffic allows platform owners to evaluate its impact before rolling it out more broadly. Traffic allocation can then be adjusted over time, and different iterations can be compared without losing historical context.

Within each scenario, the traffic split is applied consistently per session for each demand source or group of demand partners, supporting repeatable comparisons between tested configurations.

The updated A/B Testing functionality gives platform owners more control over how they validate changes before applying them across the platform. They can isolate specific traffic paths, compare configurations, and gradually increase traffic allocation based on the results.

Olga Zharuk, CPO at TeqBlaze

During an active scenario, platform owners can monitor selected metrics such as DSP spend, sRPM, and fill rate directly in the A/B Testing section. The Preview Results table also shows the configured traffic split for each scenario, while extended CSV reports are available when deeper analysis is required.

***

The upgraded A/B Testing functionality is now available out of the box in the TeqBlaze white-label SSP. To see how it works in practice, contact the TeqBlaze team to schedule a meeting.

Share this article

Stay ahead of the curve: Subscribe to our weekly newsletter