Monthly proposal economics rolled up by deal segment and region over **decided** proposals only (won + lost). `avg_margin_pct` is the average of per-proposal margin_pct; `avg_cycle_time_days` is the average decision cycle time (decision_date − submitted_date); `won_value` is realised won deal value — sumIf(total_price, is_won), where is_won = (outcome = 'won'). Pending and withdrawn proposals are excluded because their decision_date is NULL and their economics are incomplete — sweeping them in would bias margin and cycle-time averages. The gateway re-aggregates with avg() / sum() over the caller's chosen dimensions.
- highTreating these averages as covering all proposals
This metric includes decided proposals only (outcome in ('won','lost')). Pending / withdrawn proposals have a NULL decision_date and are excluded by construction — the averages describe the decided book, not the full pipeline. For still-open volume use the `rfx_pipeline` metric.
- highReading `won_value` as total submitted value
`won_value` is realised won deal value only — sumIf(total_price, is_won). Lost-proposal prices are excluded. It is NOT the total priced value across all decided proposals; pair with `proposal_win_rate` for the win conversion view.
- mediumRecomputing the averages from summed columns client-side
`avg_margin_pct` and `avg_cycle_time_days` are avg() measures. The gateway re-runs avg() over the chosen window; dividing two already-aggregated outputs across multiple months yields a sum-of-averages bias.
- “What is the average proposal margin by segment this year?”
- “How much won deal value did we close by region last quarter?”
- “What is the average decision cycle time by segment?”
- proposal_win_rate
Win conversion (won_count / decided_count) over the same decided book — pair to read margin and cycle time alongside how often proposals are won.
- rfx_pipeline
Open-opportunity pipeline (count, est_value, probability) — the upstream view before proposals are decided here.