Most subcontracting plans are written twice. Once by a capture team that treats them as a compliance annex, filling percentages against categories and naming firms nobody has spoken to. And once, eighteen months later, by a program team explaining to a contracting officer why the goals were missed. Both versions are avoidable, and the way to avoid them is the same: choose small businesses with a scored technical role, write the role so an evaluator sees capability rather than arithmetic, and use the same partners across a portfolio of bids so the relationships are real before they are needed.
This is written for the small business liaison officer and the capture director who share this problem. The plan is a required document with an evaluated substance, and the difference between a plan that earns nothing and a plan that earns credit is visible on a first read.
Why the placeholder plan fails twice
A plan built from placeholders fails at two separate moments, and the second failure is the expensive one.
It fails at evaluation because it is transparent. An evaluator who has read many plans recognizes a percentage attached to a category with no named firm, a list of firms with no described scope, or a set of goals that exactly match the agency's targets to the decimal. None of these earn a strength, and where the solicitation makes the plan an evaluated subfactor, a plan of this kind reads as risk rather than as neutral.
It fails again in performance, because the goals were never anchored to work anyone intended to give away. The program staffs the way it always intended to staff, the reporting period arrives, the numbers are short, and the explanation is a paragraph about market conditions. That conversation follows the firm into the next competition, because past compliance with subcontracting plans is a matter of record and contracting officers do look.
The plan that works is built backwards from the technical volume. Decide what a specialist small business will actually do, price that scope, then let the goals follow from the plan of record. The percentages become descriptions of a decision already made rather than aspirations.
What makes a subcontracting plan earn credit rather than pass
Editorial weighting, illustrative rather than measured. The last row is deliberately low: matching the target to the decimal reads as arithmetic, not intent.
Choosing small businesses that strengthen the technical volume
The selection criterion that produces both a good plan and a good bid is simple to state: choose firms whose presence would make the technical volume better even if no subcontracting goal existed. Everything else follows.
Depth in a narrow thing the requirement actually names. A firm that builds data pipelines and machine learning systems for federal agencies contributes paragraphs the prime cannot write, on a scope the solicitation asks about. A firm that provides general information technology support contributes a logo. The first earns a strength; the second earns a line in a table.
Delivery inside a federal environment, not just federal customers. The distinction matters. Selling to an agency and deploying inside an agency's accredited boundary are different experiences, and only the second produces the risk language and the authorization understanding that an evaluator recognizes.
People who can be named and committed. Engineers available to appear as key personnel or named technical leads, with percentages and a substitution path. A partner who cannot commit named people is contributing a capability statement.
Willingness to write. Proposal contribution is real work and it is unbilled. A partner who returns draft text in the prime's template at the assigned page count, on the prime's internal date, is worth more than a larger firm that sends a capability deck and asks what the prime needs.
A scope that survives to performance. The single best test of a proposed subcontract is whether the prime would still want it if the plan were not required. If the honest answer is no, the goals will be missed later.
Writing the role so an evaluator sees capability
The language matters more than the arithmetic, because the arithmetic is checked in a minute and the language is what produces a written finding. Three moves separate a plan that reads as capability from one that reads as compliance.
Describe the scope in technical nouns. Not participation in technical tasks, not support to the data management effort. The ingestion pipeline from the two named systems of record, the evaluation program that runs in continuous integration, the model governance documentation set. An evaluator can map those to the technical volume and see they are the same work.
Tie the role to a specific evaluated requirement. Where the statement of work asks for something, the plan should say which small business performs it and where that is described in the technical volume. Cross-references cost nothing and they tell the evaluator the two documents were written by the same team.
State what the prime gains, in engineering terms. Prior delivery of that class of system, named engineers with production experience, an existing evaluation approach, patterns already built. This is a sentence about capability rather than about socioeconomic status, and it is the sentence that turns the plan into a discriminator.
What to leave out: language that describes the small business as an opportunity or as an element of the prime's commitment to diversity. Whatever else those sentences accomplish, they do not produce a technical strength, and they crowd out the sentences that do.
The mechanics that trip teams up
A few structural facts about these plans are worth knowing before the drafting starts, because each one changes the work.
The threshold and the plan type. A subcontracting plan is required when the offeror is other than small on the applicable size standard, the award exceeds the simplified threshold set in the regulation, and subcontracting possibilities exist. The dollar threshold is adjusted for inflation periodically, so check the current figure in the solicitation rather than from memory. Individual plans cover one contract; commercial plans cover a firm's fiscal year across its commercial products and services; a comprehensive plan applies under a specific test program. Which one applies changes what the goals mean and how they are reported.
Goals are stated against subcontracting dollars, not contract value. A common drafting error. The denominator is the total planned subcontracting dollars for the contract, and getting this wrong produces goals that look generous and turn out to be unreachable, or the reverse.
Reporting is a real obligation with a system behind it. Periodic reports are filed in the government's subcontracting reporting system, and the numbers are visible to the next contracting officer. A plan written without regard to reportability produces an administrative problem every period.
Flow-down obligations reach the subcontractor. A large subcontractor receiving an award above the threshold from the prime may itself owe a plan. This matters when structuring tiers and is worth checking before a team is arranged as prime, large sub, and small sub under that.
Size status is checked against the solicitation's code. A firm may be small under one code and other than small under another, and the applicable code is the one on the solicitation. Confirm the representation is current in the government's registration system before naming a firm, and confirm it again if a year passes between team formation and award.
Where a specific number, threshold or clause reference is needed, take it from the solicitation and the current regulation text rather than from a prior bid's boilerplate. These figures move.
Building a portfolio, not a bid team
The highest-return change most large primes can make here is organizational rather than procedural: treat a small set of specialist partners as a standing portfolio rather than assembling a new team for each pursuit.
| Approach | Per-bid team assembly | Standing partner portfolio |
|---|---|---|
| Time to a signed agreement | Weeks, inside the proposal window, competing with drafting | Already executed, with a task order style addendum per bid |
| Quality of proposal text | A partner learning the prime's template under deadline | A partner who has written to it before and knows the internal dates |
| Named personnel | Resumes gathered late, commitments soft | Known engineers, prior commitments honored, substitution path understood |
| Plan credibility | Names with no history behind them | A record of goals met, described with specifics |
| Cost of the relationship | Repeated qualification, repeated negotiation | One negotiation, amortized across a year of pursuits |
| Failure mode | The partner is a name on a page and nothing more | Over-concentration; manage by keeping two or three per specialty |
The mentor-protégé structure formalizes a version of this where it fits, and the mechanics differ between the Small Business Administration's program and the Department of Defense's. Both take time to establish and neither is a substitute for the plain arrangement of working with the same specialists repeatedly until each side knows how the other operates.
The same partner across a year of bids
The plan is written per contract, but the relationship that makes it credible is annual. A specialist carried across several pursuits behaves differently from one engaged for a single bid, and the differences compound.
The first pursuit costs both sides the most. The partner learns the prime's template, its internal dates, its review cadence and the difference between what the volume lead asks for and what the volume lead needs. The prime learns whether the partner delivers text on time and whether its engineers hold up under a chief engineer's questioning. By the second pursuit, the drafting cycle is roughly half the calendar, because nobody is negotiating format.
The agreements also stop being an obstacle. A master teaming agreement with a per-pursuit addendum turns a three-week legal exercise into a one-page attachment naming the solicitation, the scope, the percentage and the exclusivity terms for that bid. Capture teams who have done this once rarely go back.
And the plan gains the sentence that matters most on the third bid: a description of goals actually met on a prior contract with this partner, with the scope and the reporting numbers named. That sentence is unavailable to any competitor who assembled a team last month, and it is the strongest thing a subcontracting plan can contain.
The management cost of this is small and specific. Keep two or three specialists per capability area so no single partner becomes a concentration risk. Review the portfolio once a year against the pursuits on the forecast. Retire the relationships that produced nothing rather than carrying names forward out of habit.
Qualifying a specialist small business in one working session
The qualification most primes run is a capability review, which tells you what a firm says it does. A better session takes ninety minutes and tells you what it can do.
- Ask them to describe a system they put into production inside an agency. The architecture, the data, the authorization path, what went wrong and what it cost. Depth and specificity in the answer are the signal; a firm that has shipped talks about failure modes without prompting.
- Ask what they would build to demonstrate the capability in eight weeks. A firm that answers with a scoped plan, a data choice and a measurement is a firm that will produce evidence. A firm that answers with a methodology will produce text.
- Ask who specifically would do the work, at what allocation. Names, percentages, current commitments and what happens if award slips. A stated constraint is planable.
- Ask them to write two pages in your template. Give the section headings, the page count and a date. What comes back tells you everything about the proposal experience.
- Ask what they assert as background intellectual property. The answer should be a short named list rather than a shrug or a claim over everything.
- Ask how they handle data and what their security posture is today. What is in place now, not what is planned.
- Ask what they would decline. A firm that says yes to every scope is a firm that will discover the boundary during performance.
Signals that a specialist small business will strengthen the bid
Editorial weighting, illustrative rather than measured. The last row is deliberately low: a firm that declines nothing has not read the requirement.
How we work as a named small business on a prime's plan
Precision Federal is a small business engineering firm. We build artificial intelligence, data platforms, software, cloud and full-stack web and mobile systems and deliver them into production inside federal agencies. On a prime's bid we take a defined technical scope, and we are comfortable being named in the plan because the scope is real work we intend to perform.
Before the proposal, we return a written technical position within a week of receiving the solicitation: what we would build, where the discriminators are, what we would demonstrate, and the risks with dates. We write our sections in the prime's template at the assigned page count, in the prime's voice, delivered on the prime's internal date rather than the government's. Our engineers attend the color team reviews and answer the chief engineer's questions directly.
In performance, the first two weeks produce the interface artifacts: the data schemas, the model artifact contract, the control responsibility matrix and the evaluation program running in the prime's own pipeline. After that both teams work in parallel with no weekly coordination tax.
The prime keeps everything that matters to the prime. The customer relationship is the prime's, exclusively, and every contact runs through it. The contract is the prime's. Code, models and definitions are assigned, with only named background tooling carved out and licensed back so nothing can be withheld. Past performance from the effort belongs to the program and the prime cites it as prime. A handover rehearsal is priced as a deliverable in each period so taking the work in-house at any option year is a staffing decision rather than a negotiation.
Pricing is either firm fixed-price increments against numeric acceptance criteria, or a committed team at a stated allocation mapped to the contract's labor categories. Reporting numbers arrive in the format the prime's plan requires, on the prime's schedule.
The first step is one email with a one-page brief: the solicitation or the portfolio you are planning, the date that matters, and the technical problem as you see it. We return a scoped, priced statement of work.
Bottom line
A subcontracting plan is evaluated on whether it describes a decision or an aspiration. Build it backwards from the technical volume: pick specialists whose scope you would want even if no goal existed, describe their work in technical nouns cross-referenced to the sections where it appears, derive the goals from the staffing plan rather than from the agency's target, and hold executed agreements before submission. Then keep the same partners across a year of pursuits so the next plan carries a record instead of a list of names. Done that way the plan stops being a form and starts adding to the score it sits beside.
Frequently asked questions
Generally when the offeror is other than small under the size standard for the solicitation's code, the anticipated award exceeds the regulatory threshold, and subcontracting possibilities exist. The dollar threshold is adjusted for inflation periodically, so take the current figure from the solicitation and the current regulation text rather than from a prior bid's boilerplate. The plan type, individual, commercial or comprehensive, changes what the goals mean and how they are reported.
Build it backwards from the technical volume. Name firms with scopes described in technical nouns, cross-reference each to the section of the technical volume where the work appears, derive the goals from the staffing plan rather than from the agency's stated target, and hold executed agreements or commitment letters before submission. A plan whose percentages match the agency target to the decimal with no named firms reads as arithmetic and earns nothing.
Choose firms whose presence would improve the technical volume even if no subcontracting goal existed. The markers are depth in a narrow capability the requirement names, delivery inside a federal accredited environment rather than merely selling to agencies, engineers who can be named and committed with percentages, and a demonstrated willingness to write proposal text in the prime's template on the prime's internal dates.
On total planned subcontracting dollars for the contract, not on total contract value. Confusing the two is a common drafting error that produces goals which look generous at submission and prove unreachable during performance, or the reverse. Since periodic reports are filed in the government's subcontracting reporting system and are visible to the next contracting officer, an unreachable goal creates an administrative problem every reporting period.
Standing partners, with two or three per specialty to avoid over-concentration. Agreements are already executed rather than negotiated inside the proposal window, the partner knows the prime's template and internal dates, named engineers are known quantities with a substitution path, and the plan carries a record of goals actually met rather than a list of names. The relationship cost is paid once and amortized across a year of pursuits.