future: github data

Problem

Fivetran's GitHub connector syncs rich engineering data — pull requests, reviews, commits, issues — but there is no pre-built dashboard pack that turns this data into actionable engineering productivity insights. Teams using the connector must build their own dashboards from scratch to answer common questions about PR cycle time, review bottlenecks, deployment frequency, and contributor activity. A GitHub dashboard pack would provide immediate value to engineering leaders who already have the data in their warehouse but lack the visualization layer.

Context

• GitHub connector data is rich enough to support meaningful engineering dashboards, but the pack catalog does not yet include a coherent GitHub productivity story.
• This is a future-opportunity task, so it should clarify the analytics narrative, likely entities, and data-model prerequisites before any pack build starts.
• The opportunity should distinguish between vanity engineering metrics and truly actionable workflow dashboards.

Possible Solutions

• A - Treat GitHub as just another connector to tackle later without planning: easy, but not strategic.
• B - Recommended: define GitHub dashboarding as a structured future opportunity: identify the key engineering questions, connector coverage, and model/content prerequisites first.
• C - Start building a full GitHub pack immediately: exciting, but too premature without clearer scope and semantics.

Plan

• [ ] Enumerate candidate GitHub KPIs (PR flow, review latency, deployment linkage).
• [ ] Define starter semantic layer/marts assumptions for the pack.
• [ ] Sketch dashboard stories for engineering leaders and team managers.
• [ ] List open research questions and external dependencies.
• [ ] Add explicit criteria for when this graduates into delivery milestones.

Implementation Progress

• GitHub issue: https://github.com/fivetran/dataface/issues/85

Review Feedback

• [ ] Review cleared