Augmented analysis

NeoLoad Web Augmented analysis uses machine learning to segment a test result into time intervals and assign key insights to each segment. It’s especially useful when you're analyzing a long test run or comparing behavior across versions.

Before you start

When you enable Augmented analysis, NeoLoad Web will perform several actions:

  • It splits the test result into intervals automatically.

  • It adds descriptions that explain system behavior in each interval.

  • It highlights performance degradation with color-coded segments.

Enable Augmented analysis

You can enable Augmented analysis directly from the Results page by following these steps:

  1. Go to the Overview tab of a test result.

  2. Select Add interval.

  3. Choose ML based auto-analysis.

NeoLoad Web generates a set of intervals and displays them as colored blocks over the timeline.

  • Green: healthy system behavior

  • Red: performance degradation

  • Blue: neutral or baseline behavior

A tooltip appears on hover and provides a short explanation of what’s happening in each interval.

Read the interval insights

Each interval shows a label summarizing system behavior along with a breakdown of contributing factors, such as:

  • Increasing user load

  • Rising error rate

  • Slower request durations

  • Fewer requests per second

Use these insights to understand performance regressions or validate improvements.

Get more from your results

Still wondering if Augmented analysis is worth using? Here’s how it can make your life easier:

  • Save time by automating the most time-consuming part of results analysis.

  • Spot issues faster by identifying patterns, anomalies, and bottlenecks.

  • Collaborate better with clear, interval-based insights your whole team can understand.

  • Speed up onboarding by helping new testers understand test behavior at a glance.

  • Track progress over time by comparing intervals across test runs.

Tip: You can also use Augmented analysis when comparing test results against a baseline. This helps you detect regressions or confirm improvements across versions.