Scoring public construction work before the team pays to chase it

Public-sector construction work is posted in public, but that does not make it easy to find. Opportunities are spread across several federal systems, each with its own search, its own data formats, and its own rate limits. Most notices do not fit. Finding the few that do takes hours every week, and pursuing the wrong one costs real money.

The client's business development team asked for five things: pull opportunities from the federal sources, match them against what the client actually builds, give each one a bid or no-bid call with the compliance details, show who has won similar work before, and run the whole thing as a pipeline with alerts.

A web app with a full phone layout, so it works from a job site.

Aggregation. Daily syncs from three federal sources: contract opportunities, award history, and grants. Grants are an early signal. The recipient usually issues the construction contract months later, so tracking grants lets the team get in early instead of competing on the open notice.

Fit scoring. Every opportunity gets a 0 to 100 score across seven weighted dimensions: industry code alignment, scope match, geography, project value, set-aside eligibility, timeline, and procurement stage. Hard disqualifiers, such as a set-aside the client cannot pursue or a passed deadline, force a no-bid. Softer ones cap the call at "consider."

Compliance extraction. Bonding, licensing, experience minimums, site visits, required forms, and wage determinations, pulled into a checklist. Where a solicitation has attachments, the app downloads the PDFs and extracts the sections covering proposal instructions and evaluation criteria.

Competitor intelligence. For each opportunity, the likely incumbent, the top competitors with their award totals, and agency spend by year, all from live federal award data.

Pipeline, alerts, and a map. Nine stages with drag and drop, alerts on new matches, amendments, deadline changes, and recompetes, and a map that plots every opportunity colored by fit, with radius search for prospecting.

Scenarios. Scoring profiles can be cloned and compared, so the team can see what changes when it pursues work alongside a certified small-business partner. Under one teaming scenario, 14 opportunities moved from no-bid to pursuable.

Competitors became their own database. Early on they were assembled on the fly, and the same firm showed up twice, once with its federal entity ID and once without. Now every competitor is a record keyed on that ID and linked to every pursuit it appears on. Before seeding, all 51 name-only firms were resolved to a real ID. "Who can prime the work we cannot" became a database query instead of a round of phone calls.

Then we checked that data against the official source. The app had been reading small-business certification status from whatever each firm claimed on its last federal award, which carries no dates. The government's own certification records showed that of 11 competitors flagged with one key certification, six had already lost it, some of them years earlier. Exit dates now refresh daily.

The most useful changes came from one client call. The teaming partner's point was that a sources sought notice is the only window where a firm can help shape a requirement. Once a solicitation posts, the scope is fixed and the field is open. The notices were already in the database, but the product buried them: there was no filter by notice type, scoring ranked early notices below posted solicitations, and alerts were gated on a score those notices could never reach. We added a capture setting that flips the stage ranking, a "shapeable window" filter, an alert type that skips the score gate, and a way to re-score stored records. We also told the team plainly that procurement stage is only 6 points of 100, so the filter does most of the work.

After re-scoring 562 records, early-stage notices went from zero in the top 25 to eight. Finding the proposal instructions section in real solicitations went from 0 of 4 to 4 of 4. The first answer on that call was that this would need dedicated hardware. It did not: the pipeline already worked in the cloud, and nothing on screen called it.

The scoring holds up because it explains itself and never contradicts itself. An early version recommended "bid" on a job it had flagged as too small, so hard disqualifiers now override the score. A first scoring pass that gave six of fourteen test opportunities a perfect 100 was rebalanced so the number actually ranks. And the federal rate limit was measured rather than read from the documentation, because the documented limit, the expected limit, and the real one turned out to be three different numbers.

Read the full build notes: A pursuit platform that scores public work before the team spends money chasing it.