Capture & Proposal · Research
The Changing Economics of Capture & Proposal Work
September 23, 2026
Government contractors have traditionally added Capture and Proposal capacity through employees, consultants, outsourced proposal support, and software. AI changes the economics because more intelligence-heavy work can now be executed without adding human labor one-for-one. The relevant question is shifting from which AI tool to buy to which work still requires human judgment, which work can move to managed execution, and what qualified work remains undone when capacity runs out.
Key findings
- Capture and Proposal are already mixed labor markets. Contractors combine internal teams, consultants, outsourced support, and software, which makes work-level substitution more practical than replacing an entire function.
- The economic constraint is capacity, not simply productivity. Qualified opportunities can be delayed, underworked, or abandoned when pursuit demand exceeds available Capture and Proposal labor.
- AI lowers the cost of intelligence-heavy work such as monitoring, research, solicitation analysis, compliance preparation, drafting, and structured review, but judgment-heavy work still depends heavily on experienced people.
- Software and managed work draw from different budgets. Software helps people perform work; managed capacity takes responsibility for producing defined work.
- The strongest business case should be measured against current labor and outsourced spend plus the value of work left undone, not against software subscription prices alone.
The research question
How does the economics of Capture and Proposal change when organizations can buy execution capacity for defined work instead of adding human labor or software alone?
Capture and Proposal have always been capacity businesses
Government contracting growth requires recurring work before and after an RFP: finding opportunities, determining fit, researching buyers and competitors, shaping pursuits, building capture artifacts, analyzing solicitations, creating compliance structures, drafting responses, and reviewing outputs. Demand for that work is uneven. Multiple pursuits can accelerate at once, deadlines compress, and the people with the most judgment are often the same people needed elsewhere in the business.
Deltek's 2025 Government Contracting Industry Study shows the pressure directly. Among surveyed contractors, 77% rated more or better opportunity identification as important, 74% cited improving sales or capture processes and technology, 62% cited earlier understanding of company fit, 56% cited investment in sales or capture staff, and 52% cited automation and AI.
Source: Deltek 2025 Government Contracting Industry Study. Percentages are Deltek survey findings, not Compound Leverage estimates.
The traditional answer is to add people
Human labor remains the reference point for the economics of pursuit work. There is no single Bureau of Labor Statistics occupation for “capture manager” or “proposal manager,” but adjacent professional roles illustrate the cost base. In May 2025, BLS reported median annual pay of $101,860 for management analysts and $110,740 for project management specialists. In professional, scientific, and technical services, management analysts had a median annual wage of $107,330. These figures are wages, not fully loaded employer cost.
Reliable internal context and judgment, but capacity is purchased whether pursuit volume is high or low.
Capacity can be added around pursuits, but cost and availability vary with scope, deadline, and expertise.
Lowers the effort required for some tasks, while the customer still owns execution and output.
Defined recurring work is carried by an external workforce while the customer retains expertise, judgment, relationships, and approvals.
APMP's 2024–2025 U.S. Compensation Report reinforces that proposal, bid, capture, business development, and graphics work represents a specialized labor market. Its findings are based on 2,269 individual survey responses. Detailed compensation data are member/licensed material, so this research does not reproduce those figures.
Outsourcing already proves the work can move outside the company
Capture and proposal support is not a new outsourcing category. Providers already sell proposal management, compliance, writing, pricing support, capture planning, and full proposal development. The market demonstrates that organizations will place defined pursuit work outside the employee base when they need expertise or surge capacity.
This matters because the transition to AI-native managed work does not require a buyer to accept outsourcing for the first time. Sequoia Capital's 2026 “Services: The New Software” thesis argues that existing outsourcing is a particularly strong wedge for AI-native services: the work is already accepted as external, a budget already exists, and the buyer is purchasing an outcome rather than a tool.
If the buyer already pays someone else to do the work, AI does not have to create a new budget. It can change how that work is produced.
Separate intelligence work from judgment work
Sequoia's framework distinguishes intelligence from judgment. That distinction is useful for Capture and Proposal because the work is not uniformly automatable. Many recurring activities are rules-heavy, research-heavy, document-heavy, or pattern-heavy. Others depend on relationships, accountability, organizational context, experience, and strategic choice.
| More intelligence-heavy | More judgment-heavy |
|---|---|
| Opportunity monitoring and intake | Customer relationships and influence |
| Initial fit and qualification analysis | Final bid / no-bid decision |
| Market, buyer, incumbent, and competitor research | Win strategy and positioning |
| Solicitation analysis and requirement extraction | Solution and technical tradeoffs |
| Compliance matrices and structured outlines | Pricing and risk decisions |
| First-draft development from approved source material | SME expertise, approvals, and accountability |
| Structured review and issue identification | Final editorial and executive judgment |
The boundary is not fixed. Model capability, source quality, workflow design, risk, and the importance of a pursuit all change where human review belongs. The operating implication is not “remove the human.” It is to reserve scarce human capacity for the work where judgment changes the outcome.
Software economics and work economics are different
A software comparison asks whether a subscription makes an employee faster. A work comparison asks what it costs to produce the required output. Those are different denominators.
This distinction is visible in the market. GovCon software products can be priced in the hundreds of dollars per month, while outsourced proposal and capture services are scoped around the work itself. Comparing a managed service only with a software license understates the relevant budget. The appropriate comparison is the labor, contractor, and outsourced capacity required to produce the same work.
The hidden variable is unworked pipeline
Capacity shortages do not always appear as an expense. Sometimes the work is delayed, performed superficially, or never started. Sequoia describes this as “abandoned work” in procurement: economically useful work that receives little attention because human labor cannot be justified at the required scale.
The Capture and Proposal equivalent is qualified pursuit work that never receives enough research, capture, shaping, drafting, or review. The pipeline value of those opportunities is not revenue and should not be treated as such. But the existence of recurring qualified work that the organization cannot adequately pursue is evidence of a capacity gap.
A better economic model
The business case for managed Capture and Proposal capacity should begin with the organization's current economics, not a generic AI productivity percentage.
| Measure | What it tells you |
|---|---|
| Current annual Capture & Proposal spend | The existing economic base across employees, contractors, consultants, and outside support. |
| Addressable work | The portion of current work that is repeatable, intelligence-heavy, and suitable for managed execution. |
| Capacity gap | The amount of qualified work that current resources cannot fully support. |
| Unworked pipeline | The contract value associated with qualified opportunities not fully worked. This is exposure, not projected revenue. |
| Managed-work fee | The actual scoped cost required to carry the defined work. |
| Projected ROI | A comparison that becomes meaningful only after scope and managed-work cost are known. |
This is the logic behind the Capture & Proposal Economics Calculator. It estimates current cost, addressable spend, capacity gap, and unworked pipeline first. A projected ROI should be calculated only after the actual work has been scoped and priced.
What this means for government contractors
The emerging decision is not “people or AI.” It is how to allocate pursuit work across internal people, external specialists, software, and managed execution. For many organizations, the most defensible starting point is recurring intelligence-heavy work that already consumes internal labor or external support. Human teams continue to own expertise, relationships, judgment, approvals, and accountability.
This changes the buying question from “Which AI proposal tool should we add?” to “What Capture and Proposal work needs to get done, who should do it, and what capacity should we own versus buy?”
Limitations
Public compensation data do not map cleanly to specialized GovCon Capture and Proposal roles, and fully loaded labor costs vary materially by employer, geography, seniority, benefits, overhead, and allocation method. Public outsourced-service pricing is fragmented and often vendor-reported. AI capability also varies by task, model, source material, workflow design, and required review. Pipeline value is not expected revenue. This research therefore treats labor and vendor evidence as directional and does not claim a universal savings rate, win-rate improvement, or ROI.
Methodology & sources
This publication combines public government-contracting industry research, U.S. labor data, professional association evidence, current market examples, and Sequoia Capital's 2026 services thesis. Vendor pricing and capability claims are used only as evidence that software and outsourced work are sold under different economic models, not as independent proof of outcomes.