Every federal R&D proposal has a Section 8 (Action Items), Section 9 (Clarifying Questions), and Section 10 (IP & Novelty). These are the sections that get drafted once, never reviewed, and skimmed by reviewers. They are also where roughly 30 percent of proposal losses originate, because they are where the team’s confusion about scope, dependencies, and IP cleanly leaks through.
The Heilmeier Questions We Blow Through
The Heilmeier Catechism has eight questions that any good R&D proposal should answer. Most teams nail the first six: what are you trying to do, why does it matter, what is new, who cares, how will you do it, how will you measure success.
Then they hit question seven, “what are the risks,” and question eight, “what has been tried before.” Both get two paragraphs of generic risk-register language with no connection to the actual technical approach. The risk matrix mentions “budget overruns” and “schedule delays” without quantifying probability or impact. The prior-work section cites three irrelevant papers nobody on the team read.
These weak answers always surface in Sections 8 through 10, where the team’s uncertainty becomes visible. A strong Action Items section forces you to translate your approach into concrete deliverables with dependencies. A thoughtful Clarifying Questions section requires admitting what you do not know about the government’s requirements or assumptions. A rigorous IP & Novelty section demands you articulate what is new versus what is incremental.
The failure mode is predictable: confident narrative sections followed by vague back-end sections that read like they were written by someone who never read the front half of the proposal.
The Sections Teams Ship Raw
The standard ten-section technical narrative template ends with three short sections that consistently get neglected:
- Action Items
- Clarifying Questions
- IP & Novelty
Most teams polish sections 1 through 7 until they gleam. The approach reads as novel, the milestones are SMART, the success metrics are unambiguous. Then they ship 8 through 10 raw. Action items list generic deliverables with no stakeholder owners or approval dates. Clarifying questions either contain softballs the team already knows the answer to or are blank. IP & Novelty recites patent boilerplate without identifying the specific claims that would differentiate this work from prior art.
The engineering details in the front sections can be brilliant and the back sections still reveal that nobody mapped the approach to concrete deliverables, nobody pressure-tested the team’s understanding of the requirements, and nobody did a prior-art search relevant to the specific approach.
Why Reviewers Read These Sections
Government program officers and technical reviewers do not read Sections 8 through 10 as filler. They read them as evidence of the team’s self-awareness about its own work.
A well-crafted Clarifying Questions section signals a team that understands the boundary between what they know and what they do not. It shows they have identified the government’s unstated assumptions and are seeking confirmation rather than making blind guesses. This is the mark of an experienced team that has survived technical projects before.
A blank or generic Clarifying Questions section signals arrogance or inexperience. Either the team thinks it knows everything about the problem domain, or it does not understand enough to know what it needs to ask. Both are red flags for reviewers who have seen the first scenario lead to misaligned execution and the second lead to project failure.
Similarly, a specific Action Items section demonstrates that the team thought through how work flows from one task to another. Generic action items suggest the technical approach has not been validated against an actual implementation plan.
The Contrarian Frame
Most proposal-coaching advice focuses on the technical sections. Teams spend weeks polishing their approach and success metrics. The actual differentiator is the meta-sections, where the team’s relationship to its own uncertainty shows up.
Strong proposals demonstrate not just technical competence but organizational self-awareness. The team that articulates its risks specifically, admits what it does not know in clarifying questions, and precisely identifies what is novel about its approach consistently beats the technically superior team that glosses over these meta-questions. The pattern holds across agencies, technical domains, and team sizes.
Review Sections 8–10 First
The practical fix is counterintuitive: review the back-end sections before you polish the front-end ones.
Most proposal reviews start at Section 1 and work forward. That is backwards. It means you polish the narrative before validating whether it holds together when translated into action items and assumptions. Instead, draft all ten sections, then circle back to 8 through 10 and use them as a validation check. If your Action Items reveal that your technical approach requires deliverables you have not planned for, fix the approach. If your Clarifying Questions expose assumptions that make your success metrics unrealistic, adjust the metrics. If your IP analysis shows your “novel” approach is incremental once you set it next to prior art, reframe the technical problem.
When the back-end sections are solid, return to polish 1 through 7. Now you are polishing a proposal that has already been validated against implementation reality, not one that reads well but falls apart when anyone tries to execute it.
What This Looks Like in Practice
A strong Section 8 lists specific action items with owners and dependencies: “Month 3: PI and government technical POC agree on test-case selection criteria,” not “Develop test cases.” Each action item should trace back to a milestone in Section 5.
A strong Section 9 asks specific questions that reveal genuine uncertainty: “The solicitation mentions integration with system X but does not specify whether the integration point is at the API level or the data level. Please confirm which integration surface is expected.” This tells reviewers you have identified a concrete ambiguity rather than admitting you have not read the requirements.
A strong Section 10 goes beyond patent boilerplate: “The novel contribution lies in the application of algorithm Y to data type Z, which has not been demonstrated in prior art [cite specific examples]. This differs from approach A [cite] through specific mechanism B, which reduces complexity by factor C.”
The Takeaway
The sections nobody edits twice are the ones where proposal teams reveal whether they understand their own work. Technical brilliance in the front sections cannot compensate for confusion in the back sections. Proposals that demonstrate self-awareness about their limitations and specific understanding of their implementation details consistently win over technically equivalent proposals that gloss over the meta-questions.
The federal proposal section no one edits twice is the section that decides who wins.