Contravalis Benchmark Query Tool

Benchmark a federal contract before you bid

Put in a contract's agency, industry and size, and see what comparable federal awards actually went for — who won them, how many firms bid, and which awards are approaching recompete.

Built for federal capture and business development teams.

Every figure traces back to a real award in the public record: USAspending.gov prime award archives. How this is built, and what it does not tell you.

Only the agency code is required. Leave NAICS, PSC or value blank and the tool lists the markets that exist in whatever you did enter — pick one from there.

Get your first pursuit brief — free

This tool answers one market cell at a time. A brief covers the agencies and NAICS codes you actually bid in: what is expiring, who holds it, how the field around it has behaved, and which awards are worth research time before a competitor gets there.

Send me your top two or three agencies, your NAICS codes and your typical contract size, and I'll build it this week.

jonas@contravalis.com
Where this data comes from, and what it doesn't tell you

The source

Public federal contract data — the USAspending.gov prime award archives, the government's own record of what it awarded, to whom, and for how much. Nothing here is proprietary, modelled or estimated: every figure is an aggregate of real award records, and the comparable awards listed below a result are the individual rows behind it.

What a comparison is made of

Awards are grouped into cells of awarding agency, NAICS industry code, product or service code (PSC) and size tier, over a recent window of fiscal years. Your estimated value picks the size tier, so a $40,000 order is compared against other small orders rather than against multi-year vehicles in the same industry.

When a cell holds too few awards for a percentile to mean anything, the observed range is shown instead of quantiles — an interpolation between three numbers is arithmetic, not a benchmark. The threshold comes from the API with each answer rather than being fixed in this page, so it cannot drift from what the analysis actually used.

The comparable awards and the recompete radar are drawn from awards above a value floor, over a recent window of fiscal years, with the radar looking a fixed horizon ahead. Those three settings belong to the build that produced this data rather than to this page, which is why you will not find them quoted as fixed numbers here — a page that repeated them would eventually be repeating them wrongly.

Where the record is incomplete

Offer counts are unreported on a large share of federal awards. The average above is computed over the awards that carry the field, and the "offer data coverage" line next to it says how many that was — an average backed by half a cell and one backed by all of it are different claims, and showing them identically would invite you to trust them equally. A low offer count is evidence about what was reported, not proof that nobody else bid.

USAspending records who won. It does not record who lost, so nothing here can tell you which firms competed against an incumbent and were not selected — only how many offers were received, where that was reported at all.

What "recompete" means here, and what it doesn't

The recompete radar lists awards whose current period of performance ends within the horizon. That is a date on the record, not an announcement: an agency may re-solicit early, extend, restructure the requirement, or not compete it at all. Ceiling growth since award is likewise a fact about modifications already made, not a prediction about the next competition.

These are research signals — reasons to spend time on something before a competitor does. They are not procurement advice, and they are not a substitute for the official solicitation. Verify any specific opportunity against the notice before acting on it.