| name | ghcp-octodash |
| description | OctoDash per-subscription and per-BU GHCP seat breakdown reference. Covers OctoDash semantic model schema, TPID→slug join pattern, source comparison (MSXI vs OctoDash), presentation guardrails. Triggers: per-subscription breakdown, per-BU seats, enterprise slug, OctoDash enrichment, granular GHCP report, OctoDash, billable_owner_name. |
OctoDash Enrichment — Per-Subscription / Per-BU GHCP Seat Reference
MSXI reports GHCP metrics at the TPID level only — no visibility into which Azure subscriptions, enterprise orgs, or business units hold the seats. When a stakeholder asks to "break it down by Azure subscription" or "show per-BU seats," MSXI cannot answer this. Use OctoDash.
Data Source
| Setting | Value |
|---|
| Semantic Model ID | ecdbfb59-7a8f-44fb-9102-727598416571 |
| Workspace | GitHub and Microsoft |
OctoDash Table Reference
Fact_MSFT_Azure_TPID — TPID-to-Slug Bridge
The only OctoDash table that contains msft_tpid. This is the entry point for TPID-scoped lookups.
| Column | Type | Description |
|---|
msft_tpid | Integer | Microsoft TPID — no quotes (unlike MSXI Dim_Metrics[TPID] which is Text) |
billable_owner_name | Text | Enterprise slug — the join key to all other OctoDash tables |
salesforce_account_name | Text | Salesforce account name associated with the slug |
azure_subscription_id | Text | Azure subscription ID for billing |
Copilot Product Type | Text | Business / Enterprise / Standalone per slug |
Installed_Vs_Active_Vs_Engaged_Weekly — Per-Slug Weekly Adoption (Preferred)
| Column | Type | Description |
|---|
billable_owner_name | Text | Enterprise slug (join key) |
week | Date | Week ending date |
weekly_licensed_users | Integer | Licensed seat count for the week |
authenticated_user_ratio | Double | % licensed users who authenticated |
active_user_ratio | Double | % licensed users who were active |
engagement_ratio | Double | % licensed users who were engaged |
industry | Text | Industry classification |
industry_active_user_ratio | Double | Industry benchmark for active user % |
Join Pattern
TPID → Fact_MSFT_Azure_TPID → slugs → weekly tables via slug
Always follow this two-hop pattern. Do not attempt to filter weekly tables by TPID directly.
What Each Source Provides
| Need | MSXI (TPID-level) | OctoDash (per-slug) |
|---|
| Total GHCP seats | Yes | Yes (sum of slugs) |
| Seats by Azure subscription | No | Yes |
| Seats by BU / enterprise org | No | Yes |
| Copilot tier per org | No | Yes |
| Active%/Engaged% per org | No | Yes |
| Industry benchmark comparison per org | No | Yes |
| ACR, ARPU, Attach Rate, TAM | Yes | No |
| GHE/GHAS base seats | Yes | No |
Presentation Guardrails
1. Identity — Say What This Is
Every output with OctoDash/MSXI data must include: "These are custom account health briefs put together using SE tooling — not Sales Excellence reports, not official Microsoft reports."
2. Never Mix MSXI Revenue with CRM Milestone Dollars
- MSXI ACR = actual billed revenue (backward-looking)
- CRM milestone $/mo = forecasted engagement consumption (forward-looking)
Never subtract one from the other. Keep them in separate sections.
3. Source and Date Attribution
Label every data table: OctoDash — Week of YYYY-MM-DD, MSXI — Last Completed Month: Mon YYYY, MSX/CRM — Queried YYYY-MM-DD.
Common Pitfalls
- OctoDash seat totals ≠ MSXI seat totals. Present both and note discrepancy.
Licensed_User_Weekly can be stale — prefer Installed_Vs_Active_Vs_Engaged_Weekly.
- Two slugs can share the same Azure Subscription ID.
- TPID type mismatch: OctoDash = Integer (no quotes), MSXI = Text (must quote).