SAM.gov and USAspending pipeline scoring for capture and BD

SAM.gov is a database, not a pipeline — and that's the problem

SAM.gov publishes federal opportunities. It does not score them, prioritize them, or tell you whether the incumbent is vulnerable. Treated as a search box, SAM.gov produces a daily list of postings the BD team has to read, filter, and judge by hand. By the time someone has read 60 postings, fatigue sets in and real opportunities get missed.

The mid-market firms winning more work have shifted SAM.gov from a search problem to a pipeline problem. Every relevant posting gets ingested, scored against the firm's capability and past-performance profile, paired with USAspending and FPDS data on the incumbent, and surfaced as a ranked queue with reasoning. The BD team works the top of the queue, not the alphabetical list of every NAICS.

Capability matching is the first scoring layer

The first score is capability fit. Does the opportunity's NAICS match the firm's primary or secondary NAICS? Does the PWS or SOW reference labor categories the firm has performed before? Does the contract type — FFP, T&M, CPFF, IDIQ — match what the firm's accounting system supports? Does the location, set-aside (8(a), HUBZone, SDVOSB, WOSB), and clearance level match the firm's posture?

We embed each opportunity's PWS and SOW text and match against an embedding of the firm's past performance, contract history, and capabilities statement. The match score is the cosine similarity, calibrated to a label set the BD team curates from won and lost bids. The output is a 0–100 capability fit score with the highest-influence text spans highlighted.

Opportunities ingested

~12K/yr mid-market BD pipeline

Capability-match top quartile

~3K/yr reviewable volume

Pursuit decisions

~280/yr after capture review

Bid win rate lift

+8–14 pts first 18 months on system

Incumbent intelligence is what turns a posting into a real pursuit decision

Every recompete has an incumbent. The capture decision depends heavily on whether the incumbent is vulnerable. USAspending and FPDS are the public data feeds that answer this: contract value trajectory (growing, flat, declining), modification history (clean or messy), small-business utilization patterns, performance period extensions or curtailments, and the contracting officer's history with the incumbent.

We ingest USAspending and FPDS daily, build an incumbent profile per opportunity, and surface vulnerability signals in the pipeline view. A clean, well-managed incumbent on a five-year mature contract is a low-probability pursuit. A messy, twice-curtailed incumbent on a recompete with a new contracting officer is a different conversation. The pipeline reflects this; the BD team's time follows the signal.

Past-performance corpus is the firm's most underused asset

Every win the firm has produced is a CPARS rating, a final report, an after-action document, and a set of staff who lived the work. Most firms keep this material in a SharePoint folder no one searches and a Word document no one updates. Capability matching needs structured access to this corpus to score against.

We build a structured past-performance store: contract metadata, CPARS ratings, lessons learned, key personnel, and the embeddings of the technical narrative. New opportunities match against this corpus and the pipeline shows which past performances are most relevant. The proposal team starts work with the right past-performance citations identified, not with a SharePoint hunt.

Agency forecasts and BD signals matter more than postings

By the time an opportunity hits SAM.gov, the agency has been planning for 6–18 months and the incumbent has been positioning. Real BD pipeline includes forward signals: agency budget submissions, GAO reports flagging the program, OIG findings creating recompete pressure, and OUSD or component-level forecasts. These don't replace SAM.gov; they precede it.

The pipeline ingests these sources where available and surfaces 'shaping opportunities' before the formal posting. The BD team gets a chance to influence requirements, position past performance, and engage the contracting office before the competition hardens. This is the part of the pipeline that actually moves win rates.

No-bid decisions are as important as bid decisions

Most firms have a bid-no-bid problem in the wrong direction: they bid too much, win less, and exhaust capture and proposal staff on losing pursuits. The disciplined BD organization no-bids most of what crosses the queue and concentrates resources on the high-probability work. The pipeline has to make no-bids easy and document the reasoning so the firm doesn't second-guess in retrospect.

Our pipeline supports one-click no-bid with reason codes: capability mismatch, incumbent strength, location or clearance, set-aside ineligibility, profitability concerns, capacity constraints. The reasons aggregate into BD analytics that show whether the firm's pursuit profile is matching its capability profile or drifting away from it.

The pipeline writes back to the GovCon ERP, closing the loop to capture and proposals

A pursuit decision in the pipeline creates a capture record in the ERP with the opportunity metadata, the capture lead, the budget, the milestone schedule, and the past performances flagged for the proposal. Capture status updates flow back to the pipeline. The opportunity, the capture, the proposal, and ultimately the contract live as a single record across the lifecycle.

This is the difference between a BD tool and a BD pipeline integrated into the GovCon ERP. The closed loop means the firm sees its win rate by capability, by capture lead, by past-performance citation, by incumbent vulnerability — and the conclusions feed forward into next year's pursuit profile.

We were bidding 110 opportunities a year and winning a quarter of them. After we tightened the pipeline and let the scoring drive pursuit decisions, we bid 60 and won a third. Same proposal staff, more work won, less burnout. Capture is the discipline of saying no to the wrong work.

— VP Business Development, mid-tier defense services firm

Frequently asked

How does the system score SAM.gov opportunities? +

It embeds the PWS and SOW text and matches against an embedding of the firm's past performance, contract history, and capabilities statement, calibrated against a label set the BD team curates from won and lost bids. The output is a 0–100 capability fit score with the highest-influence text spans highlighted. NAICS, contract type, set-aside, location, and clearance fit are scored explicitly alongside the semantic match.

What incumbent intelligence is available from public data? +

USAspending and FPDS provide contract value trajectories, modification history, small-business utilization, performance period changes, and contracting officer history. We ingest both daily and build an incumbent profile per opportunity. Clean, growing incumbents are typically low-probability pursuits; messy, curtailed incumbents on recompetes are different conversations. The pipeline surfaces the vulnerability signals.

Can the pipeline catch opportunities before they post on SAM.gov? +

Yes, partially. Agency budget submissions, GAO and OIG reports, OUSD and component-level forecasts, and acquisition strategy documents create forward signals that often precede formal SAM.gov postings by 6–18 months. The pipeline ingests these where available and surfaces shaping opportunities so the BD team can influence requirements and position past performance before the formal posting.

How does past-performance matching work? +

Every win is captured in a structured past-performance store with contract metadata, CPARS ratings, lessons learned, key personnel, and embedded technical narrative. New opportunities match against this corpus, and the pipeline shows which past performances are most relevant for each pursuit. The proposal team starts work with citations identified rather than a SharePoint hunt.

Why are no-bid decisions important? +

Because most firms bid too much, win less, and exhaust their proposal staff on losing pursuits. The pipeline supports one-click no-bid with reason codes — capability mismatch, incumbent strength, set-aside ineligibility, capacity constraints — and aggregates these into BD analytics. Disciplined no-bids concentrate resources on high-probability work, which lifts win rates by 8–14 points in the first 18 months in our deployments.

How does the pipeline integrate with the GovCon ERP? +

A pursuit decision creates a capture record in the ERP with the opportunity metadata, capture lead, budget, milestone schedule, and flagged past performances. Capture status updates flow back to the pipeline. Opportunity, capture, proposal, and resulting contract live as one record across the lifecycle, enabling win-rate analysis by capability, capture lead, past-performance citation, and incumbent profile.