Skip to content

AI Impact Analytics Metrics Matrix

Everyone can contribute. Help move this issue forward while earning points, leveling up and collecting rewards.

Duo Pro and Enterprise features, availability, metrics, and dimensions

Latest Google sheet of metrics

See archived table of metrics SKU legend:
  • 1️⃣ Duo Pro (Available for Premium/Ultimate)
    • Deployments Supported: GitLab.com, Dedicated, Self-Managed
  • 2️⃣ Duo Enterprise (Available for Ultimate)
    • Deployments Support: GitLab.com, Dedicated, Self-Managed

Metric availability legend:

  • Metric is available in internal telemetry and visible in Tableau per CRM account
  • 👁️ Metric is available in AI Impact Analytics Dashboard
  • Metric is available via GraphQL interface
  • 🚧 Metric is planned to be included in AI Impact Analytics
  • ❌ Metric is not planned to be available in AI Impact Analytics

Feature maturity:

  • GA: General Availability
  • B: Beta
  • E: Experimental
Impact Analytics Metrics Currently Supported Dimensions

code_suggestions (GA ) 1️⃣ 2️⃣
  • Unique users count
  • Unique users (%) 👁️
  • Unique users list
  • Accepted count
  • Rejected count
  • Loaded count
  • Shown count
  • Error count
  • Streams started
  • Streams completed
  • Acceptance rate 👁️
  • Median display time
  • Median load time
  • Median time to show
  • Suggestion source
  • Is supported language
  • Language
  • Delivery type
  • Extension name
  • Extension version
  • IDE name 🚧
  • Model engine
  • IDE vendor
  • Language server version
  • Model name
  • Suggestion outcome
  • Is invoked
  • Has advanced context
  • Is direct request
  • Options count
  • Response type
  • New vs. returning (user) 🚧
  • User
  • Suggestion size

duo_chat (GA) 1️⃣ 2️⃣
  • Unique user count 👁️
  • Unique users (%) 👁️
  • Unique user list
  • Questions asked count 🚧
  • Feedback count 🚧
  • Question type (General, issue question, epic question, /explain, /refactor, /fix, /tests, /troubleshoot, /summarize_comment, /vulnerability_explain, /clear, /reset 🚧
  • New vs. returning (users) 🚧

duo_chat > documentation_search (GA) 1️⃣ 2️⃣

 TBD TBD

duo_chat > categorize_duo_chat_question (GA) 1️⃣ 2️⃣

 TBD TBD

explain_vulnerability (GA) 2️⃣
  • Unique user count 🚧
  • Unique users (%) 🚧
  • Unique users list 🚧
  • Usage count 🚧
  • Resolved vulnerabilities 🚧
  • Overall Mean Time To Resolve 🚧
  • Mean Time To Resolve with Vulnerability Resolution 🚧
  • Vulnerability resolution utilization 🚧
  • Vulnerability resolution time saved 🚧
  • New vs. returning (users)

resolve_vulnerability (B) 2️⃣

Combined with explain_vulnerability

generate_issue_description (E) 2️⃣

glab_ask_git_command (GA) 2️⃣
  • Unique user count 🚧
  • Unique users (%) 🚧
  • Unique user list 🚧
  • Usage count 🚧
  • New vs. returning (users)

summarize_issue_discussions (B) 2️⃣
  • Unique user count 🚧
  • Unique users (%) 🚧
  • Unique user list 🚧
  • Usage count 🚧
  • New vs. returning (users)

generate_commit_message (GA) 2️⃣
  • Unique user count 🚧
  • Unique users (%) 🚧
  • Unique user list 🚧
  • Usage count 🚧
  • New vs. returning (users)

review_merge_request

summarize_review (E) 2️⃣

explain_code (GA) 1️⃣ 2️⃣
  • Unique user count 🚧
  • Unique users (%) 🚧
  • Unique user list 🚧
  • Usage count 🚧
  • New vs. returning (users)

generate_cube_query (E) 2️⃣

semantic_search_issue

summarize_merge_request (B) 2️⃣

troubleshoot_job (GA) 2️⃣
  • Unique user count 🚧
  • Unique users (%) 🚧
  • Unique user list 🚧
  • Usage count 🚧
  • Failed pipelines 🚧
  • Improved pipeline success ratio 🚧
  • Overall Avg. Pipeline resolution time 🚧
  • Avg. Pipeline resolution time with Root Cause Analysis 🚧
  • Root cause analysis utilization 🚧
  • Root cause analysis time saved 🚧
  • New vs. returning (users)

observability_all

Data Source: https://gitlab.com/gitlab-org/gitlab/-/blob/master/ee/config/cloud_connector/access_data.yml

AI Impact Analytics API Endpoints

Click to expand

The update information about the APIs, can be found on the AI Impact FAQ - #512931 (closed)

Edited by 🤖 GitLab Bot 🤖