professional plan

Regression Analysis

Track failure trends, flaky tests and slowing endpoints over time.

What It Does

Regression Analysis watches your test results over time to show when something starts to break. It tracks tests that alternate between pass and fail, listing each one above a 30% flip rate with its recent history, and flags tests whose recent response times are more than 50% slower than earlier runs. Recent runs are compared with earlier runs of the same tests, whatever the environment. On the dashboards edition (Enterprise and Trial), threshold alerts on pass rate, failure count, flaky-test count, p95 response time, coverage or quality score are checked every five minutes and delivered in-app, by email or by webhook. For "what broke in this run, and why", Failure Analysis on every plan groups the failures into causes with verdicts and catches regressions a green run would miss.

Run 42 of the Orders regression pack is compared with run 41 and the last eight runs: Get order is now failing, Refund order is flaky with a 38% flip rate, List orders got slower, from 120 ms to 310 ms, the rest are stable, and the flaky and slower tests are flagged with a link to the report

Overview

Spot when previously passing tests start failing, which tests flip between pass and fail, and which endpoints are getting slower. Trend analysis shows failure patterns over time, flaky tests are found by flip rate, and tests whose recent response times are more than 50% slower than earlier runs are flagged. For the explanation of why a particular run failed, Failure Analysis groups the failures into causes with verdicts. Combined with the Test Scheduler, the trends update continuously so you catch problems within hours, not days.

Key Capabilities

Historical trend analysis across test runs with visual charts
Flaky tests found by flip rate: above 30% over at least three runs
Flags tests whose recent response times are more than 50% slower
Threshold alerts on pass rate, failures, flaky count, p95 or coverage by email, webhook or in-app (dashboards edition)
Quality score: 50% pass rate, 30% consistency, 20% response-time band
Failure Analysis explains why a run failed, cause by cause

How It Works

  1. 1

    Tests run on a schedule or in CI/CD pipelines, building a history

  2. 2

    Recent runs are compared with earlier ones for pass rates and response times

  3. 3

    Flaky and slowing tests are flagged

  4. 4

    Set threshold alerts to be notified by email, webhook or in-app

  5. 5

    Open Failure Analysis on a failed run to see its causes

Available on

professional Plan & Above

Upgrade to the professional plan to unlock this feature.

Available on Professional plan and above
See pricing →

Try Regression Today

Start your 15-day free trial — no credit card required.