Write for the Evaluator, Not the Champion
I've watched a lot of small firms lose proposals they should have won, and the reason is almost always the same. They wrote a beautiful document for a reader who doesn't exist.
The document is gorgeous. Custom graphics, confident narrative, a real design sensibility. The team is proud of it, and they should be — it's genuinely impressive. And it loses to something that looks, frankly, kind of boring. This drives capable people crazy, because they're sure the better-looking, better-written proposal is the better proposal. It isn't. It's the better proposal for the wrong audience.
The winning move in competitive proposals — and this is dead center in the small-firm GovCon lane, where a five-person capture team is up against outfits ten times their size — is to stop writing for the audience you want and start writing for the one who actually decides.
There are two readers, and they are not the same person
Every serious proposal has two audiences, and confusing them is the root of most losses.
The first is the champion. The person inside the customer who wants you to win — who saw your demo, gets what you do, and is rooting for you in the room. The champion is real and matters. You win the champion with vision, with a great demo, with the exciting version of the story.
The second is the evaluator. And the evaluator is a completely different animal. The evaluator is a tired, risk-averse person sitting with a scoring sheet, a stack of competing proposals, and a finite amount of patience. He didn't see your demo. He doesn't care about your vision. He has a rubric, a limited number of minutes per proposal — often something like four — and one dominant motivation that has nothing to do with picking the best solution: he does not want to get in trouble for his choice.
That's the reader who decides. Not the champion. The evaluator, with his four minutes and his scoring sheet and his profound desire to make a defensible, low-risk call. And most proposals are written to thrill the champion while completely ignoring the evaluator, which is exactly backwards, because the champion doesn't fill out the score sheet.
Winning artifacts are engineered for his cognition
Once you understand the evaluator, a lot of things that look like "boring" proposal conventions reveal themselves as deliberate engineering. The winning proposal often looks conventional on purpose. Underneath, it's built for exactly how this tired man reads.
Mirror the evaluation language verbatim in your headings. If the criteria say the customer will evaluate "technical approach to sustainment," your heading says "Technical Approach to Sustainment" — those exact words, not your cleverer synonym for them. The evaluator is scanning for the thing on his sheet. When your heading matches his rubric word for word, he finds it in one second instead of ten, and you've just made his job easier at the precise moment he's deciding your score. Rename it "Keeping Systems Healthy for the Long Haul" and it's better prose and a worse proposal, because now he has to translate, and he's tired, and he might not bother.
Put the proof exactly where he looks. He's not reading your whole narrative to hunt for evidence you can do the thing. Put the evidence right against the requirement it satisfies, so that where he expects proof, proof is sitting there. Don't make him search. Every second of searching is a second of friction, and friction lowers scores.
Write the score justification for him. This is the one that wins. Somewhere in his process, the evaluator has to write down why he scored you the way he did — and that's work, and he's tired. If your proposal hands him a clean, defensible sentence he can essentially copy straight onto his sheet — plain language that maps your strength directly to his criterion — you've done his hardest task for him. You're not just making a claim. You're writing the justification for a good score in a form he can lift. Make it that easy and he will, because you've reduced his effort and his risk in the same stroke.
The two-audience rule
So here's how I coach small teams to think about it, and it resolves the tension between the impressive proposal and the winning one:
Wow the champion in the demo. Soothe the evaluator in the document.
Those are two different jobs for two different people, and you need both. The demo is where vision lives — that's where you win the person who'll advocate for you inside the room. Go big there. But the document isn't a second demo. The document's job is to make the evaluator's decision to score you well feel safe. Not exciting. Safe. Defensible. Easy to justify to whoever reviews his scoring.
Because that's the thing nobody says out loud about how these decisions actually get made: the evaluator isn't choosing the best solution. He's choosing the option that minimizes his personal risk. He wants a choice he won't have to defend, a call nobody second-guesses, a score he can back up in one sentence if asked. Reduce his risk — make picking you the safe, obvious, easy-to-justify choice — and you win. Dazzle him while raising his uncertainty, and the gorgeous proposal loses to the boring one that made him comfortable.
Small firms lose this game constantly because they pour their limited hours into making the proposal impressive, which serves their own pride and maybe the champion, and they underinvest in making it safe to choose, which is the only thing the deciding reader actually responds to. The bigger competitors, the ones with real proposal shops, figured this out long ago. It's not a resource advantage. It's an understanding advantage, and it's available to a five-person team the moment they stop writing for the reader they wish they had.
Write for the evaluator. He's the one holding the pen.
The same rule works outside government
I've been talking about GovCon because that's where the two readers are most formal — there's an actual scoring sheet and an actual rubric. But the pattern isn't about government at all. It shows up in every competitive proposal a small business writes, because there are almost always two readers.
Think about a commercial bid. The champion is the person who brought you in, who liked your pitch, who wants this to work. The evaluator is the CFO who's never met you, scanning the proposal for the reason to say no, worried about defending the spend to a board. Or the committee member who wasn't in your demo and is comparing three bids on price and risk. Different titles, same two jobs: someone to inspire, and someone to reassure. And the losing move is identical — pour everything into impressing the person who already likes you, and hand the skeptic nothing to hold onto.
So even without a Section M to mirror, the discipline transfers. Find out who actually signs off, and what would make their decision feel safe. Mirror the language they'll use to justify the choice. Put your proof where their eyes will land. Write the sentence they'll repeat to whoever asks them why they picked you. The formality changes; the fact that the deciding reader is choosing the safe option, not the best one, does not. That's true in a federal source selection and it's true in a homeowner choosing between three contractors. Write for whoever holds the pen.
If your team competes on proposals and keeps losing the ones you were sure you'd win, the problem may be who you're writing for. Here's how we work.
You might also like
The Proposal Math — Why Most Service Businesses Leave Money on the Table
A good proposal takes 10-20 hours. Your team can run 3-4 per month. That means you're passing on opportunities every month — not because you can't win them, but because you can't write them.
The Capture Capacity Trap: Why Mid-Tier GovCon Firms Can Only Pursue 10 Deals a Quarter
APMP benchmarks show each federal proposal costs $15K-$40K in labor. Mid-tier GovCon firms max out at 8-12 pursuits per quarter — and lose 75-85% of them. AI agents can 10x pursuit volume without adding a single capture manager.
Building PropelAI: Why I Built an AI Product After 20 Years of Selling Services
I've spent twenty years in government technology. I've been a Deputy CIO. I've run federal practices. I've led captures worth half a billion dollars. And in almost every war room, the same scene plays out.