> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mergeguide.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Benchmarks methodology
> How MergeGuide compares your metrics against an industry baseline.
The **Benchmarks** area of the dashboard compares your security and quality metrics
against an industry baseline and shows where you land as a percentile. This page
explains how those baselines are derived and how to read the comparison.
## What you see
For a chosen category and period, the benchmarks view shows:
* **An overall percentile** — where your metrics land relative to the baseline.
* **A per-metric table** — your value, the baseline median, your percentile, a
status, a **confidence label** (LOW / MEDIUM / HIGH), and a recommendation.
## Where the baseline comes from
Baseline percentiles are derived from **published industry research**, including:
* BSIMM14
* Veracode State of Software Security v14
* Sonar State of Code Quality 2023
* OWASP SAMM Benchmark 2022
* GitHub Octoverse 2023
* GitLab Global DevSecOps Survey 2023
* GitGuardian State of Secrets Sprawl 2024
* Qualys TruRisk Research 2023
* Tenable Time to Remediate 2022
* Verizon DBIR 2024
## How to read it
* The baseline is an **estimated baseline derived from published research, not an
empirically measured value** for your peer set. The view labels it as such.
* Each metric carries a **confidence label** so you can weight it appropriately.
* Individual metrics vary with your industry, team size, and development practices —
treat the comparison as directional guidance, not a precise ranking.
Benchmarks are an industry-grounded estimate to help you orient, not a
customer-specific guarantee. Read each metric alongside its confidence label.
## Next steps
Where benchmarks live.
How hours-saved is estimated.