mart_sme_coverage
certifiedWhich experts are loaded into the pipeline, who is available, and who contributes to wins.
Per-SME coverage view. Joins the RFX subject-matter-expert directory (department, title, seniority, region, expertise tags, certifications, weekly capacity, availability) to two pipeline rollups: `assigned_opp_count` (opportunities where the SME is the owner) and `authored_proposal_count` / `won_proposal_count` (proposals where the SME is the lead author, and of those how many had outcome `won`). Use it to answer staffing questions ("who has headroom in EMEA security?") and win-contribution questions ("which authors win the most?"). Snapshot, not a time series. `full_name` is PII (name) and is masked for PII-excluding personas via column grants.
- mediumRanking authors on raw `won_proposal_count`
`won_proposal_count` is an absolute count, so prolific authors rank highest regardless of quality. For win quality, divide by `authored_proposal_count` (guard against divide-by-zero — it is 0 for SMEs who authored nothing).
- mediumTreating a zero count as missing data
`assigned_opp_count`, `authored_proposal_count`, and `won_proposal_count` are coalesced to 0, not null — the mart LEFT-joins from the directory so an SME with no pipeline activity is a real 0, not a gap. Don't filter `IS NOT NULL` to find "active" SMEs; filter `> 0`.
- mediumReading `full_name` from a PII-excluding persona
`full_name` is PII and is excluded from `ai_readonly` / `rfx_contributor` via column grants. Selecting it as those personas returns a permission error, not a null. Use `sme_id` as the join/display key for those personas.
- “Which available SMEs have security expertise in EMEA?”
- “Who are the top win-contributing lead authors?”
- “Which SMEs are loaded into the pipeline but unavailable?”