Every newly ingested opportunity goes through a rule-based scoring engine and comes out with one of three verdicts. There is no model in the path and no black box: eight signals, each worth a fixed number of points, and a total you can decompose and defend.
How the score is computed
By default, eight signals combine into a total with a maximum of 14 points:
| Signal | Max points |
|---|---|
| Title matches your lane keywords | 2 |
| Description matches your lane keywords | 1 |
| Agency tier (federal, tier-1 state, and so on) | 2 |
| Win-theme keywords and legacy-incumbent flags | 2 |
| Past-performance NAICS match | 2 |
| Procurement vehicle (GSA eBuy, NASPO, Sourcewell, and others) | 2 |
| Estimated seat count | 2 |
| Recompete detected | 1 |
Every scored opportunity keeps a receipt of which rules fired and what each contributed, so any verdict can be explained line by line.
The three verdicts
- Pursue: the total is at or above your Pursue threshold. The opportunity moves to Scored, contact enrichment runs, a CRM deal is created, and the contact is enrolled in a sequence.
- Partner: the total lands between your Partner and Pursue thresholds. The same downstream actions fire, but the opportunity is flagged for teaming partners rather than a direct bid.
- NoBid: the total is below the Partner threshold. Downstream stages are skipped. The opportunity is kept in your workspace for reporting, but no contact is created and no email goes out.
Tuning the rules
Thresholds, rule points, and keyword lists are all editable on the scoring rules page in your workspace settings, where you can also add rule types beyond the starter set, including set-aside category, contract-value thresholds, and regex patterns. Changes apply to opportunities scored after the change; anything already scored keeps its original verdict. To re-score a single opportunity after a rule change, remove it and let the next ingestion run pull it back in.