What does PWin mean in government contracting?

PWin stands for Probability of Win. It is a percentage between 0 and 100 that estimates the likelihood your company wins one specific bid, based on how you score against the factors that drive the award decision: customer relationship, competitive position, technical fit, past performance, key personnel, price position, teaming, compliance risk, capture maturity, and operational readiness. It is a decision aid for allocating bid and proposal budget, not a prediction of the outcome.

What is a good PWin score?

Most capture organizations bid above 70, treat 50 to 69 as conditional on retiring specific named risks, and no-bid below 50 unless there is a concrete path to move a factor. The exact gates matter less than setting them in writing before the meeting and calibrating them against your own win history. If your average predicted PWin is far above your actual win rate, your ratings are inflated rather than your gates being wrong.

How is PWin calculated?

The standard method is a weighted scorecard. Define 8 to 12 factors, assign each a weight reflecting how much it moves an award decision in your market, rate each 1 to 5 against a written rubric, then compute the weighted average and normalize it to a percentage. In Excel that is SUMPRODUCT of weights and ratings divided by 5, divided by the sum of the weights. Some teams then apply a calibrated logistic transform so mid-range scores spread out more realistically. CLEATUS publishes a free PWin Calculator that runs this exact 10-factor model in the browser.

Can AI calculate PWin automatically?

Yes, and the input side is what changes. CLEATUS calculates a PWIN automatically on every matched opportunity, from the capabilities, past performance, and preferences in your capture profile plus your real award history, computed against the pre-shredded solicitation package rather than a form somebody filled in. Ask the agent and it works the number deeper, researching the incumbent and the competitive field and showing every driver. It also scores against your own bid/no-bid rubric if you prefer yours. The go/no-go decision stays human.

Is PWin the same as price-to-win?

No. PWin is the probability that you win. Price-to-win (PTW) is the target price your bid needs to hit to be competitive. They inform each other, since a strong technical position at an uncompetitive price is not a strong position, but they answer different questions and should be tracked separately.

Can we keep our existing bid/no-bid gates and workbook?

Yes. Upload your existing Excel model to the Document Hub and CLEATUS scores solicitations against your factors, weights, and scoring rules, including multi-tab workbooks. Your governance gates, approval thresholds, and color review process stay exactly as they are. The platform changes who does the reading and evidence gathering, not who makes the decision.

How do amendments affect a PWin score?

Every amendment can move factors, which is why rescoring after Q&A and addenda is a core discipline. Manually this means rereading the changed sections and rerunning the sheet. In CLEATUS the amended package is reread as part of the same Contract Breakdown and the affected factors update, so you see what moved and why rather than rebuilding the analysis.

Does an automatic PWin score replace capture judgment?

No. It replaces the reading, the data gathering, and the triage, which is where capture teams lose most of their time. The judgment calls, whether to commit B&P money, which risks are acceptable, what the win strategy is, stay with people. The value of automating the score is that it runs on every opportunity rather than the handful you had time to evaluate by hand.

What Is PWin? How to Calculate Probability of…

Probability of Win (PWin) is the number that decides where your bid and proposal budget goes. Most capture teams still produce it the way they did in 2005: a spreadsheet, a conference room, and a set of ratings nobody can trace back to evidence. Here is the model that holds up under scrutiny, the free calculator that runs it, and what changes when the score arrives before anyone opens the opportunity.

TL;DR

PWin is the probability you win one specific bid, expressed 0 to 100. It is not your historical win rate, and it is not price-to-win. It exists to allocate money.
The working model is a weighted scorecard. Eight to twelve factors, each weighted by how much it moves an award decision, each rated against an anchored statement rather than a feeling.
The gates most teams use: 70 and above is a strong bid, 50 to 69 is winnable with focused capture work, below 50 is no-bid territory unless you can move a specific factor.
Run the model free. The PWin Calculator is this exact 10-factor scorecard, with an evidence-quality setting that turns the output into an honest range instead of a false-precision point.
The expensive part was never the arithmetic. It is reading the package and gathering the evidence, which is why most teams only ever score the handful of pursuits they had already decided to chase.
CLEATUS calculates your PWIN automatically on every matched opportunity, from your capture profile and real award history, computed against the full solicitation package. No questionnaire, no forms, no file upload.

What PWin Means, Exactly

PWin is the probability that your company wins one named pursuit, at one point in time, given what you know today. It is a decision aid for capital allocation, not a prediction of the award. A 65% PWin does not mean you will win. It means that if you bid twenty pursuits that look like this one, you should expect to win roughly thirteen, and that a bid here is a defensible use of the budget.

Three things get confused with PWin constantly, and keeping them apart is most of the discipline:

PWin is not price-to-win. Price-to-win (PTW) is a target price: what your bid has to cost the government to be competitive. PWin is a probability. They inform each other, since a strong technical score with a misaligned price is not a strong position, but they answer different questions and should never share a cell.

PWin is not your win rate. Win rate is historical and aggregate, measured across many bids after the fact. PWin is forward-looking and specific to one pursuit. Your win rate is useful mainly as a calibration check on your PWin model, which is a point most teams skip.

PWin is not a fit score. Plenty of tools score how well an opportunity matches your NAICS codes, your set-aside status, and your past work. That is qualification, and it is genuinely valuable for triage, but a complete PWin also weighs the competitive field, the incumbent, your customer relationship, and your price position.

The reason to hold the definition tightly is that PWin becomes a number in your forecast. Multiply PWin by contract value and you get weighted pipeline value: the only pipeline figure a CFO should be willing to look at. A $12M pursuit at 25% is $3M of expected revenue and should be resourced like $3M, not like $12M.

Why PWin Decides Where Your B&P Money Goes

A disciplined PWin process is not about being right on any single bid. It is about the portfolio:

It puts B&P hours where they convert. Proposal time is the scarcest resource in a small or mid-sized contractor, and it is spent in whole units. Every hour on a 20% pursuit is an hour not spent on a 65% one.

It makes the forecast credible. A pipeline weighted by PWin survives a board meeting. A pipeline reported at raw ceiling value does not survive the first quarter it misses.

It exposes strategic gaps you would otherwise argue about. When Past Performance scores a 2 on eleven pursuits in a row, that is not eleven bad opportunities. That is a message about teaming, recruiting, or where you should be spending internal research and development money.

It ends the loudest-voice problem. A written model with published rating rubrics moves the bid/no-bid meeting from "I have a good feeling about this one" to "show me the evidence behind that 4."

How to Calculate PWin: The 10-Factor Weighted Model

At its core, PWin is a weighted average. You define the factors that influence an award decision in your market, assign each a weight, rate each on a 1 to 5 scale, and normalize.

The factors that actually move an award

Most working models cover some version of these ten. Weights should reflect your strategy and, more importantly, your buyer's stated evaluation emphasis:

Customer and Relationship: access, intimacy, whether you helped shape the requirement
Competitive Landscape: incumbent advantage, known competitors, real discriminators
Technical Fit: solution compliance, innovations, retired risks
Past Performance: relevance, recency, CPARS quality
Key Personnel: named resumes, availability, labor mix
Price-to-Win and Cost Realism: position against budget, wrap rates, risk
Teaming and Small Business Strategy: partners, socioeconomic plan, workshare
Compliance and Proposal Risk: RFP interpretation, volume plan, color reviews
Capture Maturity: gate artifacts, customer calls, Black Hat complete
Operational Readiness: facilities, tools, clearances, transition plan

Start the cost-realism review with the free wrap rate calculator. It separates direct labor, indirects, and fee before the bid/no-bid review.

A worked example

Rate each factor 1 to 5, where 1 is weak and 5 is outstanding, and weight by strategic importance. The normalized score is your initial PWin.

Factor	Weight (W)	Rating (R, 1-5)	Weighted = W*(R/5)	Notes
Customer Relationship	12	4	9.6	We've met PM; requirements validated
Competitive Position	10	3	6.0	Incumbent likely rebidding
Technical Fit	15	5	15.0	Strong solution; minor risks retired
Past Performance	12	4	9.6	2 highly relevant CPARS
Key Personnel	10	3	6.0	1 key resume still contingent
Price-to-Win	15	3	9.0	Target within budget; refine wraps
Teaming Strategy	8	4	6.4	SB partner locked; 25% workshare
Compliance & Risk	10	5	10.0	Full L/M traceability plan
Capture Maturity	8	3	4.8	Black Hat in progress
Operational Readiness	10	4	8.0	Transition draft complete
TOTAL	110	-	84.4	-

Normalized score in Excel: =SUMPRODUCT(B2:B11, C2:C11/5)/SUM(B2:B11) gives 0.767 (76.7%)

Many teams use that normalized score directly. Advanced teams apply a calibrated logistic transform so mid-range scores spread out more realistically instead of clustering around the mean:

=1/(1+EXP(-8*(NormalizedScore-0.5)))

Calibrate the 8 and the 0.5 against your own historical wins and losses. Do not copy someone else's constants.

Skip the spreadsheet. This exact model runs as a free tool. The PWin Calculator asks you to pick the anchored statement that is true today rather than an abstract number, tags each rating as assumed, partially evidenced, or cited, and returns a PWin range plus the three capture moves that raise the score fastest. Everything runs in your browser and nothing is uploaded.

Calibrate against your own history

Here is the check almost nobody runs. Pull your last twenty bids, record the PWin you assigned at the bid/no-bid gate, and compare it to what you actually won. If your average predicted PWin was 68% and you won six of twenty, your model is not broken, your ratings are inflated. Tighten the rubrics until a 4 requires evidence a stranger would accept. A model that predicts optimism consistently is worse than no model, because it launders the optimism as arithmetic.

How to Read the Number

Thresholds should be written down before the meeting, not negotiated during it:

70 and above: bid. Strong position. Confirm price-to-win alignment and lock the teaming agreements.
50 to 69: conditional. Winnable, but only if named risks are retired. Write down which ones and who owns them by when.
Below 50: no-bid or team. Pursue only with a concrete path to move a specific factor. "We will work harder on the proposal" is not a path.

Pair every PWin with a confidence level, because a 72% built on assumptions is a different decision than a 72% built on cited evidence. A simple rule: confidence is high only when every factor rating cites a source, whether that is an RFP section, a CPARS excerpt, or a dated customer call note. If half the ratings are guesses, the number is a hypothesis and the real output of the meeting is a list of things to go find out.

Where Manual PWin Breaks Down

The mechanics look like this, and most capture leads will recognize every step:

Collect the basics. Key the solicitation number, NAICS, set-aside, period of performance, and estimated value into Bid-No-Bid Calculator.xlsx.
Decode the package. Read Sections L and M, the PWS or SOO, exhibits, and attachments, then translate requirements into your model's factors.
Score by committee. Route the sheet around. Five people apply 1 to 5 ratings with five different interpretations of what a 3 means.
Tweak the weights. Someone nudges a weight, and now two versions of the truth exist.
Assemble rationale. The capture lead spends hours adding comments and links so leadership trusts the number.
Redo it after every amendment. An L/M change lands on Friday. Return to step 2.

Good teams still spend hours to days reaching a number everyone mostly trusts, right before the actual proposal work starts. But the slowness is the small problem. The structural problem is what the slowness causes.

You only score the pursuits you already decided to look at. Scoring is expensive, so it gets rationed. A team that can run ten PWin exercises a quarter will run them on ten opportunities that were selected by hand, informally, from a title, a NAICS code, and a set-aside on a search results page. The most consequential filtering decision in your capture process, the one that determines which pursuits ever get evaluated at all, is made with the least information and no model behind it. Everything downstream is a rigorous process applied to a pre-narrowed list.

That is the part worth fixing. Not the arithmetic, which Excel has handled since 1987.

PWin Discipline: Eight Practices That Hold Up

Define factors and scales once. Publish rubrics stating what a 3 and a 5 mean. Anchored statements beat adjectives.
Weight what your buyer weights. Mirror the Section M evaluation emphasis. Do not overweight what your team happens to be proud of.
Gate with artifacts. Require proof before a high rating: call notes, a signed teaming agreement, a draft org chart.
Avoid double counting. If transition risk is scored under Technical Fit, do not score it again under Compliance and Risk.
Keep price honest. Coordinate PWin with PTW. A 5 on Technical with a misaligned price is false comfort.
Calibrate quarterly. Compare predicted PWin against outcomes and adjust weights, rubrics, or both.
Rescore after every amendment. Q&A responses and addenda move factors. Record the delta and what caused it.
Show your work. Short rationales and document links are what make leadership trust the score instead of relitigating it.

What Changed: A PWin Score You Do Not Have to Calculate

Everything above is still the right model. What has changed is who does the reading and when the score appears.

CLEATUS calculates your PWIN automatically on every matched opportunity. There is no questionnaire, no form, and no file to upload. The score comes from what the platform already knows about you: the capabilities, strengths, and past performance in your capture profile, your real award history, your geography, and the Must Have, Prefer, Avoid, and Must Avoid criteria you set once. It is produced before anyone on your team opens the record.

It is computed against the whole package, not the cover sheet. Every matched opportunity is pre-shredded on arrival. CLEATUS reads every page (the base solicitation, amendments, attachments, and exhibits) and breaks it into one standardized Contract Breakdown following the Uniform Contract Format: scope, requirements, deliverables, deadlines, pricing, and evaluation criteria. Your score reads Sections C, H, K, L, and M plus the attachments, rather than a title, a NAICS code, and a set-aside.

That timing is the whole difference, and it is worth being precise about it. Other tools analyze a solicitation when a user clicks analyze, opportunity by opportunity, which means the analysis only ever runs on pursuits a human already selected. Pre-shredding runs on every match, which means the triage decision itself is informed.

When the package has not been read, the platform says so. If a solicitation's documents are not available, the recommendation card is marked as having limited contract details and the confidence drops rather than the model inventing a judgment from a title. A score that admits what it does not know is the only kind worth putting in a forecast.

Layer two: ask the agent to work the number deeper

The automatic score is triage. When a pursuit clears triage and deserves real capture attention, ask for the deeper analysis in GovCon Copilot:

"Work the PWin on this one. Who is the incumbent, what has this office awarded them, how does our past performance compare, and where does our price need to land?"

The agent researches the competitive field and the incumbent, pulls the contracting office's award history, searches the web for current wage determinations and market rates, and shows every driver behind the number instead of returning a bare score. That is the analysis a capture manager would spend two days assembling, and it is grounded in verifiable sources you can click.

Or stop asking and put it on a trigger. The same analysis ships as a workflow template, the PWin Agent. It fires on every new match above a score threshold you set, reads the package, researches the incumbent and the competitive field, scores the ten weighted factors (or your own rubric, if it is in the Document Hub), and writes the PWin memo. Pursuits at or above your bid gate land in the pipeline with the memo attached. Everything below comes back with the two or three risks that held the score down. Nobody has to remember to run it.

Layer three: your own rubric, and the meeting artifact

If your organization has a bid/no-bid workbook refined over fifteen years, you keep it. Upload it to the Document Hub and ask CLEATUS to score a solicitation against your factors, your weights, and your scoring rules. It reads the sheets, factors, and weights, fills the ratings in, and cites both the RFP section and the company record behind each rating. Your governance gates do not change.

And for the bid/no-bid meeting itself, where a room of people has to agree on a number and own it, use the manual worksheet. That is what the free PWin Calculator is for: a shared, anchored scorecard that produces a defensible range and a copyable summary for the minutes.

The three layers answer different questions, and a mature capture process uses all three:

Layer	Question it answers	Effort	When
Automatic PWIN	Is this worth a human's attention at all?	None	On arrival, every match
Agent deep dive	Who else is bidding, who holds it now, and where does our price need to land?	One question	When a pursuit clears triage
Your weighted rubric	Do we, as a company, commit B&P money to this?	A meeting	At the bid/no-bid gate

Manual PWin

✕Someone reads the whole package before any score exists
✕Ratings vary by whoever filled in the cell that week
✕Spreadsheet sprawl and two versions of the weights
✕Every amendment restarts the analysis
✕Only the pursuits you already chose ever get scored

With CLEATUS

Every matched opportunity arrives pre-shredded and scored
Ratings tied to cited RFP sections and your company record
One place to look, with the drivers shown
Rescored on the new package, with the change visible
Triage runs on everything, so nothing good gets filtered out by a title

From Opportunity to Decision: What This Looks Like

Monday. Three opportunities land in your pipeline already read and already scored. Two come back low against your profile with the reason attached: one requires a facility clearance you do not hold, the other is a recompete in a region you marked Avoid. Nobody spent a morning discovering that. The third scores high on technical fit and past performance relevance.

Monday, ten minutes later. You ask the GovCon Copilot to work the third one: incumbent, the office's award history, your comparable past performance, price position. It comes back with the incumbent named, four years of task orders under the same office, your two most relevant contracts mapped against the PWS, and a flag that your proposed program manager is not yet committed.

Wednesday. The bid/no-bid meeting runs the weighted worksheet with the evidence already assembled. The score lands at 0.71 with two open risks: the contingent PM and a price-to-win sensitivity around the budget ceiling. Bid, contingent on locking the PM.

Friday. An amendment lands. The package is reread, the affected factors move, and you see what changed rather than rebuilding the sheet.

The work in that week is judgment. The reading is not.

Implementation: What to Set Up in the First Hour

Load your company record properly. Past performance write-ups, CPARS excerpts, capability statements, key personnel resumes. The automatic score is only as good as what it knows about you, and this is the single highest-return setup step.
Set your Must Have and Must Avoid criteria. Deal-breakers belong in the profile, not in your head. A clearance you do not hold or a geography you will not staff should suppress a match, not surface it.
Bring your existing workbook. Rename factors clearly, make sure weights sum to a round number, and upload it to the Document Hub.
Write one-sentence rubrics. One line per rating level is what prevents rating inflation. Do this once and it pays out on every pursuit.
Set thresholds and owners. Decide in advance who can green-light an exception when PWin lands at 0.49.
Recalibrate quarterly. Compare predicted PWin against actual outcomes and adjust.

Frequently asked questions

Stop Scoring Only the Pursuits You Already Chose

The weighted PWin model is sound, and you should keep it. What is no longer defensible is spending your best capture hours reading packages so a spreadsheet can be filled in, while the opportunities you never had time to open go to someone else.

Let the score arrive with the opportunity. Spend the meeting on the decision.

Get Started →

Free GovCon Tools

PWin Calculator: the 10-factor scorecard from this guide, with evidence-weighted ranges and ranked capture moves
AI FAR Navigator: search and read the Federal Acquisition Regulation with citations
Capability Statement Builder: build a one-page capability statement
GovCon Glossary: plain-language definitions for the acronyms in your solicitation
All free tools

Customer Stories

About CLEATUS

CLEATUS is an agentic AI platform that helps government contractors discover the right opportunities, manage capture pipelines, and write winning proposals. Automate your GovCon operations end-to-end: build custom AI-powered automations that handle multi-step processes on autopilot, a force multiplier for your capture and BD team that works in the background while you stay in the loop on every decision. We aggregate federal, state, local, and city opportunities; our GovCon Copilot analyzes solicitations and your internal documents to deliver actionable market intelligence that drives revenue growth.

What Is PWin? How to Calculate Probability of Win for Government Contracts

TL;DR

What PWin Means, Exactly

Why PWin Decides Where Your B&P Money Goes