How to Write a Winning Proposal
- Deirdre Gamill-Hock
- Jul 26
- 14 min read
Updated: 2 days ago
A proposal has one job: win the business.

Not “describe the product.” Not “impress the reader.” Win it.
That goal shapes every choice I make, from the first read of the RFP to the last edit before submission.
Winning takes more than clean, compliant writing. It means reading past the requirement to the problem underneath it. It means shaping the answer so it's the one an evaluator remembers after reading the next ten. Compliant and well-written gets you in the running. Opportunity-focused and memorable is what gets you chosen
It's easy to treat each RFP as its own puzzle, solved from scratch under deadline pressure. Quality depends on how much time is left and who's available to write. That produces uneven work, and evaluators notice.
The voice shifts. The pieces never quite assemble into a polished, finished picture. A proposal that's sharp in section two and sloppy in section five reads like it was rushed at the end. Usually, it was.
Why a Repeatable Process?
Starting from scratch invites guesswork, and guesswork costs time. A repeatable process ends both. It frees that time for what matters: polishing the response into one finished, cohesive document.
I write RFP responses that are clear, compelling, and chosen using the same method every time. The process isn't a list of preferences.
It’s what turns a stack of technical requirements into a document an evaluator wants to keep reading.
Six Steps to a Winning Proposal

Proposal responses may be dense and technical, but they still need to connect with the reader. The reader may be AI ticking boxes, an SME hunting for details, or a CEO skimming for the big picture.
Reviewing my submissions with a critical eye taught me something. I noticed a pattern: items under my control that I overlooked. I was checking every box but not engaging the evaluators.
Six different ways to lose a reader:
A vague win theme
Writing that never changes pace
Dense language
A forgettable shape
A point the evaluator skims past
Leaning on AI too hard.
On the right side of the page are the six building blocks that I now use to ensure that I don’t miss a fixable problem that can turn my proposal into a failed bid.
Stack them in any order, but leave one out, and the proposal may fall apart during its evaluation.
Where It Starts: Win Themes and Pain Points

You pick up the RFP and shred it. You break down every requirement and action item until you know exactly what to do, and how to do it. That's step one, and every proposal writer gets there.
But shredding only gets you so far. What is the opportunity really worried about? A requirement is rarely just a requirement. It's usually a scar left by a vendor who missed a deadline, blew a budget, or disappeared after signing.
Find that scar. Speak to it directly.
That's where win themes come from.

A Real Win Theme Has Four Parts
A win theme has four parts.
The customer's problem or pain point
A solution that’s specific to your company rather than generic to the industry
An outcome backed by evidence
Repetition throughout the response.
Somewhere in the solicitation, usually between the lines, is a story about what went wrong before.
The win theme answers that story directly, instead of supplying facts that only meet the requirement.
A Win Theme Isn’t A Slogan.
It’s the specific reason this evaluator, with this particular problem, should pick you over the next submission in the stack.
Finding it means reading past the requirement to the risk behind it.
Why does this agency care so much about transition timelines?
Why is there an entire section devoted to reporting cadence?
Three Considerations For Your Win Theme
Make a Difference
A win theme needs more than a good pitch.
Does your win theme have:
A differentiator (something your company does better)
A discriminator (something few competitors can do)
If you can’t prove your product is better or unique, there isn’t a win theme. You just have features with good intentions.
Better can mean faster speed or newer technology.
Unique can mean an advantage that rivals seldom match.
Where's Your Proof?
A win theme without evidence is just a claim.
Evaluators don’t score confidence. They score proof. If you say you’ll cut onboarding time, show the number from the last contract where you did it.
For example, If you say your staffing model reduces turnover, cite the retention rate.
The outcome is what turns “we’re good at this” into something an evaluator can check a box next to.
What Was That Again?
The theme should not just show up once in a technical narrative nobody rereads.
It has to show up everywhere: in the executive summary, the section introductions, and the graphics. An evaluator who only sees it once won’t remember it was ever there.
The win theme analysis of the RFP and the opportunity sets the foundation for every other step. Skip it, and even a technically flawless proposal reads like a compliant, complete, and forgettable form letter. The evaluator finishes it having learned nothing about why this bidder is different from the other six proposals on their desk.
Anchor + Amplify

Proposal Evaluation Happens in Layers
When an opportunity, or its representative, evaluates a proposal, it isn't one person reading it once, start to finish.
Evaluation happens in layers: sometimes a lone reader, sometimes a full team, each layer checking for something different. Anchor + Amplify writes to every layer at once, not just the one that happens to read first.
Anchor + Amplify Satisfies Every Pass
How are proposals evaluated? Industry research points to a multi-pass process, and each pass looks for something different.
A fast first pass sorts submissions. It asks: does this belong in the stack?
An AI-assisted review tool may scan for structure, keyword alignment, and requirement coverage. It scans: does this clear the automated check first?
A closer content review checks understanding against the evaluation criteria. It checks: did they answer the question we asked?
A final comparative pass scores competitors against each other. It decides: who proved it best?

Every Response Has An Anchor
A skimmable anchor is a clear, direct answer a human evaluator can absorb in seconds. Even on a fast first pass through a stack of submissions.
The anchor states compliance explicitly and up front, in the RFP’s language. This allows the response to be scored correctly, even if nothing else in the gets read.
Underneath it sits the amplify layer.
It carries the structured, detailed content that satisfies a careful second read by a technical SME. It can also give AI-assisted scoring tools the explicit, well-organized language they’re built to reward.
The Two Layers Stay Separate
Compliance and narrative don’t share a sentence on purpose. If they did, a human reader could get a bare fact with no reason to care about it.
An automated pass could miss the compliance claim buried inside a story. The anchor proves the box is checked. The amplify layer proves it matters.
Don’t write your proposal for just one audience.
Passing the AI review matters.
Engaging the human reader, at both a skimmable level and a deeper one, matters just as much.
Anchor + Amplify writes for both.
Plain Writing
Write to be Understood

Write for the audience. That’s the rule. Your proposal may be read by a CEO, a technical lead, or an AI scoring tool. But how do you do that?
Every reader in the room absorbs your writing differently: some skim, some dig in, some run out of patience halfway through. The Federal Plain Language Guidelines are the shortcut that reaches all of them anyway.
Some people think plain writing means dumbing down your language or writing style. Instead, it means using language your audience already understands. It means organizing content so the information they need is easy to find and follow.
Plain writing isn’t about impressing evaluators with long sentences, big words, or confusing syntax. It’s about using your skills to make your win theme easy to see, so your proposal stands out and wins.

Quick tips for plain writing
Prioritize the active voice.
Use “we” and “you.”
Keep sentences to one main idea, 20 words or less.
Limit paragraphs to one topic.
Replace complex vocabulary with common, everyday words.
Eliminate unnecessary filler words.
Remove jargon.
Format so content is easy to scan.
Storytelling Frame

Does your proposal start with “Once upon a time”? No? Good.
Storytelling in proposal writing isn’t about writing scenes. It’s about making sure your technical responses don’t read like a textbook written by a 90-year-old academic.
Every section, paragraph, and sentence should have a real narrative arc: beginning, middle, end.
First, tell the evaluator what they need to know.
Then build the case with evidence, method, and specificity.
Finally, close the loop by stating the outcome plainly.
Why does it work?
Readers store new information by linking it to what they already know. A story format gives them more connection points than isolated facts.
There’s a reason this works beyond habit or preference: narrative transportation. It’s what happens when a story absorbs you enough to change how you process it. It even changes how you recall the information later.
Decades of research have found the same pattern. A structured narrative gets remembered and accepted more readily than a flat list of facts.
Whether it's a technical requirement response or explaining a differentiator, give it the same shape a good story has:
a beginning that sets up the problem
a middle that works through it
an end that resolves it
Don't Bury the Lede
Journalists use this method and even have a name for when you forget it: "burying the lede." It's when the most important fact in the story sits three paragraphs down. The reader has to keep reading, working for something they wanted from the first line.
Compare Two Openings to the Same Requirement.
Buried: “Our firm has decades of experience managing utility relocations on transportation projects of varying size and complexity, and we have developed a range of internal processes to support schedule adherence.”
Lede: “We complete utility relocations in the first 60 days, removing the single biggest risk to an 18-month schedule.”
The lede version is shorter and says more, because it puts the outcome first and lets everything else support it.
Bury the outcome, and the evaluator has to search for it, page after page. Lead with it instead, and the evaluator already has the answer before turning the page. Multiply that gap across a dozen proposals a day, and bullet twelve disappears among hundreds of others just like it.
The proposal that survives is not the one with the most bullet points. It's the one that tells a coherent story about how the problem gets solved.
Lead with the outcome, and the evaluator's brain creates a container to hold the information. The details of how your product or service will solve their problem have somewhere to attach instead of floating loose. Now every fact has a hook to hang on, and a place to be found again.
Burying the lede will frustrate the reader, and they'll lose interest before they understand the story you're telling. Start with it, and that's what the evaluator remembers the next day.
Make it Memorable: Mild Alliteration, Visual Elements and Formatting
Mild Alliteration
I know what you're thinking.
"Isn't that a poetry thing, and why would I use it?"
An alliteration is a repeated consonant sound, like "Peter Piper" or "brand-new buyer."
The repetition isn't just decoration. A matched sound catches the ear before the brain even processes the words. It's what makes the phrase stick in memory, not just the meaning.
Used once, at the right moment, that stickiness works for you. Overuse it, and the ear catches every phrase, not just the one that matters. The effect disappears the moment it becomes a pattern instead of a surprise.
Effective: “Our measured milestones keep the project on track.”
Ineffective: “Our precise, proactive, proven partners provide powerful, practical performance on every project.”
Use it right:
Catchy titles
Memorable benefits
Strong action steps
Keep it rare
Don’t let it backfire:
Too much sound
Loss of Meaning
Wrong Tone
Why Does It Work?
Cognitive research on processing fluency shows a pattern: information that's easier to say gets trusted more, not just remembered more. A phrase with matched sounds takes less effort to process. The brain reads that ease as a signal of truth, even when it has nothing to do with the claim itself.
This is why alliteration shows up in slogans, political phrases, and, yes, proposals. It isn't a trick. It's a shortcut the brain already uses, and skilled writers use it on purpose instead of by accident.
Visual Elements and Formatting

Visual Elements Give the Eye Somewhere to Land
Proposal writers are trained to read every word. You know who probably isn’t trained to do that? CEOs, technical experts, AI readers. Basically, everyone looking at your carefully prepared proposal. To be fair, evaluators skim by necessity, not by choice. They have too many pages and too little time.
What Do Readers Read?
Readers sweep across the top of a page first.
Then they sweep across a second time, a bit lower, covering less ground.
Then they scan straight down the left edge.
The shape looks like the letter F.
The middle of paragraphs and the right side of the page mostly go unread.
Remember that the goal is to make it easy for the evaluator to pick your proposal for the win.
You may have the best product, the sharpest win theme, and airtight proof. Don’t ignore presenting it in a way that makes it easy for the reader to understand and digest.
Visual elements and formatting give a skimming reader somewhere to land.
Use The “F” Pattern To Draw In The Reader
Put the point on the left, first.
Lead with the anchor sentence, the win theme, or the number that matters. Don’t bury it mid-paragraph where the F-pattern never reaches.
Eye-tracking studies confirm it: readers hit the first two lines fully, then trail off fast.
Use headers the reader can scan top to bottom.
A skimming eye reads headers before it reads anything else.
A header that states the point (“We reduce onboarding time by 40%”) does more work than one that just labels a topic (“Onboarding Approach”).
Give the eye somewhere to land.
Callouts, tables, and bolded key phrases catch a skimming reader who’d otherwise miss the point entirely.
A limited use of color can add a pop that draws attention to important points.
Pick one font and stay with it.
A single, consistent font for headers and body text lets a skimming eye recognize structure without reading a word.
It also lets the reader focus on the content instead of the formatting.
Please room on the page.
White space matters as much as the words do. The eye needs to rest so the reader isn’t overwhelmed.
Single spacing and small margins subconsciously signal more work before the evaluator even starts reading.
AI-Assisted With Human Judgment
AI can help your proposal win.
Or it can be the reason you lose.
The difference is entirely in how you use it.
Most people can recognize AI writing. And, consciously or subconsciously, they don’t like it. But we can’t ignore tools like Microsoft Copilot, which can save hours of tedious work.
You might think “Good, I can respond to more RFPs.”
But the real benefit is that it gives you time to work on the little details that matter. More time for an SME review. More time to create callout boxes. More time to hone your win theme.
AI can get a proposal 80% of the way there. The last 20% must come from you. The win theme that lands. The sentence that proves you understand this evaluator’s problem.
AI helps, but your critical thinking and writing skills are the part that set up the proposal to win.
What AI Is Good At
Analyze complex documents.
Compliance mapping and matrix building.
Brainstorming win themes.
Turning a blank page into a working outline
Repeating structure (such as Anchor + Amplify methodology) across sections.
Automating repetitive manual tasks including editing and proofreading functions.
What AI Can't Do
Proposal workflow
Strategic differentiation
Relationship nuance
Quality control and fact-checking
Ownership and accountability
Judgment and creativity still land squarely on the human writer
What to Watch for When Using AI
AI writing has tells. Once you know them, you can't unsee them.
Neither can an evaluator who's read a hundred AI-drafted proposals this year.
An AI agent will pull from incorrect or outdated information from your source library.
Hallucinations: AI will confidentially generate false, misleading, or entirely fabricated information.
Obvious AI-written prose.
Buzzwords: delve, leverage, empower, seamless, etc.
The Rule of Three: listing things or grouping descriptive adjectives in sets of three.
Formulaic Contrast: Predictable structural phrases such as “It is not just X, it is also Y” or “No fluff. No filler. Just results.”
Cliché Openings: Starting articles or essays with phrases like “In today’s fast-paced digital landscape” or “As the world continues to evolve.”
Overly Diplomatic Tone: Remaining hyper-neutral, balanced, and eager to please without taking a strong, messy, or genuinely emotional human stance.
Predictable Conclusions: Wrapping up text with rigid summary transition words like “In conclusion,” “Overall,” or “Ultimately” followed by a neat recap.
Punctuation Habits: Heavy reliance on em dashes (—) for dramatic pauses or explanatory clauses, combined with a lack of natural sentence fragments or casual imperfections.
Wondering how to get that 80% done by AI?
It starts with the prompt.
Consider AI an inexperienced writer. You have to tell it exactly what you want, in more detail than feels natural. The prompt will be longer than you think, and that's fine. Save it once it works. If you're using Copilot, turn a good prompt into an Agent so you're not rebuilding it from scratch every time.
Where To Find prompts
Feed the AI real examples of your past work such as a strong win theme or a section you're proud of. Asking it to match that voice outperforms describing the voice in the abstract.
Does your company have a style guide? Attach it and tell your AI to analyze and use it.
Ask the AI to help you write the prompt itself. Try, "What do I need to tell you to create this?"
Shortcuts and Tips
Everything I've written about here can be used to create prompts. All AI tools I use, Responsive.io, Copilot, Gemini, and Claude, contain prompts or agents exactly as I've outlined.
Tell the AI to write in a human voice, not an AI voice.
Tell it what tone to use: constructive, persuasive, or whatever fits.
When the AI gets something wrong, correct it and say why: "This needs to sound more direct, not corporate."
A Patient Tool
It's easy to feel like you're being rude when you tell AI to redo something. It is human nature, even though we understand we are using a tool. But it won't take feedback negatively and asking for another version costs nothing.
Be as direct as you need to be to get the result you want. Claude and I have discussions where we push back on each other and "argue." I have to admit; it can create a better product.
Warning
Do not feed proprietary client information, pricing, or past-performance specifics into an AI tool without knowing where the data goes. Free tools usually mean shared information, so check the fine print or use a company-approved portal.
Why It All Adds Up
Write to Win, Not Just to Submit
Not everything that creates a winning proposal is under the writer’s control. Which makes it that much more important that we optimize what we do control. In other words, we have to craft a document that reaches every evaluator – human and AI.
The six steps of my method are what’s yours to control. Master them, and every proposal gets its best shot.
Put these pieces together, and the shape of the strategy becomes clear:
Read for the real problem
Answer for skimming and AI as well as deep dives.
Include why they should care.
Write to make reading easily digestible and informative.
Shape the answer for impact and memorability.
Never let AI have the last word.
Writing clear, compelling and chosen proposals is not easy, ut applying a reliable methodology like mine can help. A repeatable process to stack the blocks instead of juggling them will save you time and effort.
Giving you more time to write more winning proposals.

Key Digital Sources & Guidelines
This post brings together key strategies from my own lived experience as a proposal writer, professional seminars, training, and reading related books and articles. For items with linkable material, I have listed the sources below.
Plain Language Action and Information Network (PLAIN), Federal Plain Language Guidelines — https://www.plainlanguage.gov/media/FederalPLGuidelines.pdf
Digital.gov, “Principles of Plain Language” — https://digital.gov/guides/plain-language/principles
U.S. Office of Personnel Management, “Plain Language” (overview of the Plain Writing Act of 2010) — https://www.opm.gov/information-management/plain-language/
Arphie, “Mastering RFP Evaluation: Essential Strategies for Effective Proposal Assessment” (multi-pass evaluator review process) — https://www.arphie.ai/articles/mastering-rfp-evaluation-essential-strategies-for-effective-proposal-assessment
Green, M. C., & Brock, T. C. (2000). “The Role of Transportation in the Persuasiveness of Public Narratives.” Journal of Personality and Social Psychology, 79(5), 701–721. Discussed in Thomas, V. L., et al. (2024), “Narrative Transportation: A Systematic Literature Review and Future Research Agenda,” Psychology & Marketing, 41(8), 1805–1819 — https://onlinelibrary.wiley.com/doi/full/10.1002/mar.22011
Dyson, M. C., et al., research on skimming and comprehension, summarized in “One Page of Text: Eye Movements During Regular and Thorough Reading, Skimming, and Spell Checking,” PMC — https://pmc.ncbi.nlm.nih.gov/articles/PMC7198234/
Cleat.ai, “AI for Government Proposal Writing: A Practical Guide (2026),” citing the 2025 Deltek Clarity GovCon Study on proposal drafting time and AI adoption — https://www.cleat.ai/blog/ai-proposal-writing
GovEagle, “Can Government Agencies Detect AI in Proposals? What GovCon Teams Need to Know in May 2026” — https://www.goveagle.com/blog/can-government-detect-ai
Nielsen, Jakob (2006, updated 2017), “F-Shaped Pattern For Reading Web Content,” Nielsen Norman Group — https://www.nngroup.com/articles/f-shaped-pattern-reading-web-content-discovered/
APMP (Association of Proposal Management Professionals), “How to Shred an RFP” — https://www.apmp.org/Web/Web/Learning-Resources/Article_Center/How_to_Shred_an_RFP.aspx


Comments