PROGRAM

Benchmark

The planned Benchmark will let your team compare results with similar delivery environments using approved aggregates. Local reporting runs on your own systems. Participation is voluntary and separate from report generation. Aggregate comparison begins once a dataset exists.

How your data stays private

The CEED Specification defines reporting from local CEED Change Passports. An implementation such as the CEED Router generates an inspectable report on your own systems, with no network calls during generation. CEED Delivery Contracts, CEED Change Passports, source code, file paths and change-level records stay inside your organization. Taking part in Benchmark requires a separate decision.

For planned participation, first prepare an aggregate contribution file locally under a private pseudonym. Inspect the file, destination and digest, then decide whether to send it. Local reporting and sharing remain separate steps.

Why participate

If your team chooses to contribute, its approved aggregates help build a useful comparison for everyone using the CEED Framework.

See your results in context

Compare your team's block rate with similar delivery environments. The context can help you choose which differences to investigate.

Help find what actually works

Comparing similar environments can point to questions worth investigating, such as why one cohort blocks more changes. Participation is self-selected, so differences are observations, not proof that a control caused an outcome.

Always voluntary, never a scorecard

Participation is opt-in, and results are never used to rank or score individuals or teams.

Register interest in participation →

Two signals, one question

That comparison starts with the control measures in the current contribution format. A future extension could relate them to independently measured delivery outcomes.

Control signal

The current contribution format contains defined CEED Change Passport-native merge metrics and declared context.

Delivery signal

A future extension can compare these control measures with independently measured delivery outcomes such as deployment frequency, lead time and change-failure rate.

As the dataset grows, participants can examine whether control measures and delivery outcomes move together in similar environments. The result identifies questions to investigate.

Build the comparison dataset

Voluntary aggregate contributions would create the first Benchmark dataset. Contributors could then compare their own control measures with similar delivery environments.

Understand the surrounding model

See how the CEED Router generates the records Benchmark would observe, or review the separate Conformance program.