What the 2025 federal procurement data tells us about RFP volume
SAM.gov publishes the official record of federal solicitations and contract awards. Here is what the 2025 dataset says about RFP volume and what proposal teams should read into the trend.
The federal government publishes its procurement record. SAM.gov carries active and archived solicitations; USAspending.gov carries award data tied to those solicitations. Together they are the highest-quality dataset on federal RFP volume that exists, and they are open for anyone to query.
This post is a read of what the 2025 record shows about federal RFP volume, with a deliberate restriction: we are not publishing specific year-over-year percentage shifts in this post. The reason is methodological. SAM.gov’s posted-solicitation count and USAspending.gov’s awarded-contract count measure different things, both have known data-cleaning quirks, and the year-over-year comparison numbers depend on which subset (full-and-open, set-asides, IDIQ task orders, simplified-acquisition) you include. Anyone publishing a single headline number without naming the slice is publishing a number without a denominator.
What we can say from the public dataset, with confidence, is the shape of the trend. The shape is what proposal teams should be reading.
Volume continues its drift upward
Federal solicitation posting on SAM.gov has been on a slow upward trajectory through the post-pandemic period. The 2025 record extends that trajectory rather than breaking it. Categories that produced large volume increases in earlier years — IT modernization, cybersecurity, cloud services, professional services contracts — continued to produce above-average volume in 2025. Categories tied to facilities, maintenance, and basic supplies remained steady year over year.
The drift is not dramatic. Federal procurement volume is not “exploding” or “doubling” or any of the framings vendor blogs sometimes adopt. It is rising at a rate that compounds across multiple years into a meaningful capacity problem for vendor proposal teams who haven’t grown headcount proportionally.
The category-mix shift matters more than the headline number. A team that was responding to mostly fixed-fee professional services RFPs in 2020 is, on the 2025 trend line, responding to a higher mix of cybersecurity-flagged and cloud-services-flagged solicitations. The compliance language in those subcategories is denser. The required certifications (FedRAMP, CMMC, FISMA tiers) are more specific. The capture work is heavier per RFP because the buyer’s evaluation criteria are more elaborate.
Set-aside volume held its share
Federal small-business set-asides, 8(a) contracts, HUBZone designations, and women-owned and service-disabled-veteran-owned small business set-asides continued to make up a meaningful slice of 2025 solicitation volume. The exact proportion shifts by category and by quarter, and the precise percentage depends on whether you count solicitation count or solicitation dollar value (these can move in opposite directions). The qualitative read: the set-aside infrastructure of federal procurement is working as designed, and small-business vendors continued to see comparable opportunity in 2025 to recent years.
For vendor proposal teams, this means socioeconomic certifications continue to be a meaningful component of capture strategy, and the “should we file for X certification” question is worth re-examining at the start of every fiscal year.
Task orders against existing IDIQs continued to dominate the high-velocity end
A pattern that has held for years and continues in 2025: a large fraction of federal procurement activity is task-order issuance against existing indefinite-delivery, indefinite-quantity vehicles. GSA Schedules, GWACs, agency-specific BPAs and IDIQ contracts. Task orders are smaller-scope, faster-cycle solicitations issued against a pre-qualified pool of vendors.
For proposal teams, the task-order pattern produces a different shape of pipeline. Fewer giant 50-page responses; more 10-to-15-page task-order responses on tighter deadlines, often with a 5-to-10-day turnaround. Win rates on task orders against an IDIQ you hold tend to be much higher than win rates on full-and-open solicitations — the field is smaller, the qualification work was done at IDIQ award time, and incumbency effects are stronger.
A vendor that is on the right vehicles is responding to a different RFP world than a vendor that isn’t. The 2025 data extends that pattern; it doesn’t reshape it.
What the data does not say
We want to be careful about three things the published dataset does not directly tell you.
Win rates. SAM.gov and USAspending.gov tell you about solicitations and awards. They do not tell you the number of vendors who responded to a given solicitation, which means a vendor-level win-rate calculation is not derivable from the public data. APMP’s annual benchmarks survey practitioners directly to estimate win rates. The federal record does not.
Proposal-team capacity. The data tells you the buyer’s volume of solicitations. It does not tell you the seller’s volume of responses, or the seller’s response capacity per FTE. VisibleThread’s writing on government proposal teams has covered this ground qualitatively — proposal teams are working harder per RFP because the documents are longer and the compliance language is denser — but the procurement record itself doesn’t quantify that.
Quality of fit. Volume is not opportunity. A vendor whose ideal-fit category produces 20 RFPs a year has the same opportunity in a year that produces 100,000 federal solicitations and a year that produces 110,000. The macro number is interesting; the per-vendor read is what matters.
What proposal teams should read into 2025
Two things, in order.
First: the trend is enough to make capacity a real planning question. If you responded to 30 federal solicitations in 2024 and your capture team is producing the same kind of pipeline in 2025 with the same headcount, you are working harder per RFP because the documents are denser. That cost compounds. Right-sizing capacity — by automation, by content reuse, by tighter bid/no-bid discipline, by all of these — becomes a financial decision rather than a workflow preference.
Second: the category mix matters more than the headline volume. Cybersecurity and cloud-services RFPs require certifications that take 6 to 18 months to obtain. A team waiting until they see the right RFP to start the certification clock will lose the RFP. The 2025 trend rewards teams who started those certifications in 2024.
How to query the dataset yourself
Both SAM.gov and USAspending.gov expose APIs and downloadable datasets. SAM.gov’s API surfaces active solicitations with category, set-aside designation, and posting date. USAspending.gov’s data covers awarded contracts with vendor name, award amount, and NAICS code.
A vendor proposal team that wants to understand its own opportunity space — which categories are growing in their NAICS, which agencies are issuing more in their service area, which vehicles they should be on — can pull that data without a vendor or a subscription. The data quality varies (USAspending.gov has known categorization gaps), but for a directional read, the public dataset is the right starting point.
We will publish a follow-up post on a specific NAICS category later this year, with the methodology written up in enough detail to be reproducible. For now, the read above is what the trend says without committing to a specific numerator-over-denominator the post can’t fully defend.