Methodology · version 1.0.0

How the RFP Detector Score works

How attractive and workable is this RFP for a typical qualified bidder, today? One number from 0 to 100, the same for every visitor, with every reason shown on the RFP page and in the report.

Starts at 506 weighted factorsSame for everyone, never for sale

Open RFPs by grade, live9,911 scored open RFPs · median 54

Excellent 80–100213 · 2%
Strong 65–792,139 · 22%
Fair 50–644,092 · 41%
Limited 35–492,665 · 27%
Poor 0–34802 · 8%

The badge colour follows the grade; the letter and the label always say the same thing in text.

What the score measures

The RFP Detector Score answers one question: How attractive and workable is this RFP for a typical qualified bidder, today? It rewards real money on the table, open and winnable competition, enough time to respond, a clear and complete notice, an issuer that actively buys this kind of work, and newness.

Every RFP starts at 50. Six factors move it up or down, and the RFP page lists each contribution with a sentence explaining it. The contributions add up to the score.

Grades and labels

The number maps to a letter grade and a label. The badge colour follows the grade; the letter and the label always say the same thing in text.

Excellent

80–100

213 open now · 2%

Strong

65–79

2,139 open now · 22%

Fair

50–64

4,092 open now · 41%

Limited

35–49

2,665 open now · 27%

Poor

0–34

802 open now · 8%

Live distribution across 9,911 scored open RFPs; median score 54. A notice capped for not being openly competed is labelled “Not competed”.

The formula

Every RFP starts at 50. Each factor adds or subtracts points depending on how this RFP compares with a typical one (the prior), scaled by its weight and by how sure we are of the data. Unknown data contributes exactly 0. The page shows every contribution, and they sum to the score.

score = clamp( round( 50 + Σ K × wᵢ × cᵢ × (sᵢ − pᵢ) ), 0, 100 )

K = 1.8 (stretch constant) · w = weight · c = confidence (0–1) · s = sub-score (0–1) · p = the sub-score of a typical RFP. Without K, six independent factors would put most RFPs between 40 and 60.

The six factors

Weights are relative influence. “Typical” is the sub-score of an average RFP, so a factor only moves the score when this RFP is better or worse than typical and we know it.

Factor weights
Factor	Weight	Typical	Most it can move	Question
Value	20	0.5	−18 / +18	How much money is on the table?
Competition & access	20	0.5	−18 / +18	Who can bid, and how concentrated is the market?
Issuer & market	20	0.65	−23.4 / +12.6	Does this issuer actively buy this kind of work?
Time & stage	15	0.55	−14.9 / +12.2	Is there a workable window to respond, and is this a real chance to bid?
Clarity & completeness	15	0.7	−18.9 / +8.1	Is the notice clear and complete enough to respond to?
Freshness	10	0.5	−9 / +9	How new is it?

Value

weight 20

How much money is on the table?

Inputs

Published value or range
Budget stated by the buyer
Federal: median contract award above $25K for the same NAICS and agency (federal award history)

Measured: The official notice or the buyer published a value.
Estimated: Federal: an award-history bucket with at least 20 awards. Other RFPs: the median published value of similar RFPs in the same state (at least 10 values).
Unknown: Nothing usable. Scored as neutral (no effect).

Competition & access

weight 20

Who can bid, and how concentrated is the market?

Inputs

Set-aside or eligibility restriction
Sole-source signals in the notice
Share of dollars won by the largest recipients (federal award history)

Measured: The set-aside is known and market concentration comes from an agency-level award bucket.
Estimated: The set-aside is known, but concentration comes from an all-agency bucket or is missing.
Unknown: The notice states no eligibility restrictions (common on web RFPs). Scored as neutral.

Issuer & market

weight 20

Does this issuer actively buy this kind of work?

Inputs

Contracts awarded and dollars obligated for the NAICS and agency (federal award history)
Other RFPs from the same issuer on RFP Detector in the past 12 months

Measured: An agency-level federal award bucket.
Estimated: An all-agency bucket, or RFP Detector history for non-federal issuers.
Unknown: No award history and no issuer activity yet. Scored as neutral.

Time & stage

weight 15

Is there a workable window to respond, and is this a real chance to bid?

Inputs

Days until the response deadline
Notice stage (solicitation, presolicitation, sources sought, RFI, special notice)

Measured: Always for federal notices and buyer-posted RFPs (no deadline is itself a fact).
Estimated: A web RFP whose deadline was extracted with less than 80% confidence.
Unknown: A web solicitation with no deadline found. Scored as neutral.

Clarity & completeness

weight 15

Is the notice clear and complete enough to respond to?

Inputs

Description length
Number of documents
Contacts (email, phone, named people)
Codes and place of performance
Solicitation number and deadline

Measured: At least 90% of the components are known.
Estimated: Some components are unknown (for example, a description we have not retrieved).
Unknown: Never fully unknown: contacts and codes are always known.

Freshness

weight 10

How new is it?

Inputs

Days since posting
Days since the last material amendment (counts up to 80%)

Measured: Always.

Curves and inputs

Each factor turns its inputs into a sub-score between 0 and 1. These are the exact curves the scoring code uses.

Value

Log scale: $25.0K → 0, $10.0M or more → 1. A range uses its geometric mean; “up to $X” counts as 60% of X.

Sub-score by contract value

View as table

Contract value	Value
$25.0K	0
$100.0K	0.23
$300.0K	0.41
$1.0M	0.62
$3.0M	0.8
$10.0M	1

Time & stage

Days until the response deadline. Capability-statement notices (sources sought, RFI, special notice) need less time. The result is multiplied by the notice stage below.

Bids (solicitations)Capability statements

View as table

Days left	Bids (solicitations)	Capability statements
0	0	0
1	0.05	0.1
3	0.2	0.4
7	0.5	0.8
10	—	1
14	0.8	—
21	1	—
60	1	1
120	0.85	—
365	0.7	0.8

Freshness

Days since posting. A material amendment refreshes the factor, counting up to 80%.

Sub-score by days since posted

View as table

Days since posted	Freshness
0	1
2	1
7	0.75
14	0.5
30	0.25
60	0.1
90	0.05

Clarity & completeness

Weighted components. Web RFPs are scaled by extraction confidence.

Description35%
Documents25%
Contacts20%
Codes and place10%
Identifiers10%

Notice stage multiplier

Notice stage multipliers
Notice	Multiplier
Solicitation (RFP, RFQ, IFB)	× 1
Combined synopsis/solicitation	× 1
Call for proposals	× 1
Presolicitation	× 0.85
Sources sought	× 0.75
Request for information	× 0.75
Special notice	× 0.6
Bundling notice	× 0.6
Justification (J&A)	× 0.3
Award notice (not scored live)	× 0
Other notice	× 0.6

Competition: who can bid

Set-aside pool scores
Set-aside code	Pool
Full and open competition NONE	0.45
Total small-business set-aside SBA	0.85
Partial small-business set-aside SBP	0.7
8(a) competitive set-aside 8A	0.95
HUBZone set-aside HZC	0.95
Service-disabled veteran-owned small business set-aside SDVOSBC	0.95
Women-owned small business set-aside WOSB	0.95
Economically disadvantaged WOSB set-aside EDWOSB	0.95
Veteran-owned small business set-aside (VA) VSA	0.95
Local area set-aside LAS	0.9
Indian small business economic enterprise set-aside ISBEE	0.95
Indian economic enterprise set-aside IEE	0.95
Buy Indian set-aside (IHS) BICiv	0.95
8(a) sole source 8AN	0.05
HUBZone sole source HZS	0.05
SDVOSB sole source SDVOSBS	0.05
WOSB sole source WOSBSS	0.05
EDWOSB sole source EDWOSBSS	0.05
Veteran-owned sole source (VA) VSS	0.05

A narrower eligible pool scores higher for those who qualify. Sole-source codes mean there is effectively no competition to enter.

The typical federal contract award above $25K is about $68.7K (median of 1,322,079 new awards, FY24–FY26).

Measured, estimated, unknown

Measured: taken directly from the notice or the buyer.
Estimated: inferred from award history or similar RFPs, so it moves the score less.
Unknown: no data, so the factor has no effect.

Unknown never counts against an RFP: a factor we know nothing about contributes exactly 0 points.

Data confidence

Data confidence = Σ weight × confidence: how much of the score rests on measured data.

High: 75–100Medium: 50–74Low: 0–49

Caps, closed and cancelled RFPs

Not openly competed (justification for other than full and open competition, sole-source set-aside codes, "intent to sole source" or brand-name-only notices): capped at 30 and labelled "Not competed".
Purely informational notices (option exercised, award notification, cancellation): capped at 30.
Closed, awarded and cancelled RFPs show the score at publication, frozen; they never show a live score.

The score at publication is computed as of the day the RFP was posted, so a closed page never shows a score dragged down by an expired deadline. The badge turns grey and reads “Score at publication”.

Score at publication Not scored yet (never shown as F)

What the score is not

It is the same for every visitor; personal relevance is the separate Fit score shown to signed-in organizations.
It does not estimate the probability of winning and does not rate the issuer as an organization.
It is informational and is not advice to bid or not bid. No user, organization or payment data feeds it, and nobody can buy or edit it.

Score vs Fit

Fit (0–100) shows how relevant an RFP is to your organization’s matching profile. It is private to signed-in organizations and never changes the public score.

Fit weights
Fit component	Weight
NAICS	35
Keywords	25
Eligibility	15
Geography	15
Contract size	10

Fit caps
Fit is capped when	Max
Title matches an excluded keyword	20
Requires a certification you have not listed	25
Not openly competed	30

Labels: Strong fit 75+ · Possible fit 50+ · Weak fit 0+. Fit is only visible to signed-in members of the organization.

Version and changelog

Every score stores the version that produced it. A change to any weight, curve or rule gets a new version and a dated entry here, and open RFPs are rescored overnight. Closed RFPs keep the version they were frozen with. Current version: 1.0.0.

v1.0.02026-11-12
- First public version: 50 plus the signed contributions of six weighted factors, stretch constant K = 1.8.
- Grades A ≥ 80, B 65–79, C 50–64, D 35–49, F < 35; notices not openly competed or purely informational are capped at 30.
- Federal value estimate = median federal contract award above $25K for the NAICS and awarding agency, last three full fiscal years.

Inputs: Public procurement records (federal, state, local, education, private and non-profit) · Federal award history (federal RFPs) · RFPs posted on RFP Detector by buyers. See how it works for how RFPs are collected and when they last updated.

FAQ

Why is this RFP's score low?

Open the breakdown on the RFP page. Each factor shows its points and a sentence: common reasons are a short response window, no published value, a sole-source signal, or a notice with little description and no documents.

Can a buyer change their score?

No. Buyers can improve clarity by publishing complete information: a full description, documents, contacts and a clear deadline. Nobody can pay for or edit a score.

Is the score personalised?

No. It is the same for everyone. The separate Fit score is personal to your organization and only visible when you are signed in.

How often is it updated?

Open RFPs are rescored when they change and every night, because time to respond and freshness change daily. Closed RFPs keep their score at publication.

Should I bid because the score is high?

The score is informational and is not advice to bid or not bid. Use it to triage, then read the notice and confirm every detail with the issuer.

RFP Detector is an independent service and is not affiliated with any government agency. Scores are informational.