Writing proposals is one of the most time-consuming parts of the sales process — and one of the areas where the gap between high performers and everyone else is most visible.
Most salespeople either spend too long on proposals (four, six, sometimes eight hours on a single document) or they use the same generic template for every prospect and wonder why their win rate is stuck. AI eliminates both problems. It removes the hours without sacrificing the personalization that wins deals.
Here's the exact workflow I teach.
Why Proposals Take So Long (And Why They Don't Have To)
A proposal has four core components: the problem statement, the proposed solution, the investment, and the why-us. Of those four, only one requires truly original thinking every time — the problem statement, because it has to reflect what you actually heard in discovery. Everything else can be templated, adapted, and AI-assisted.
Most salespeople write everything from scratch every time. That's where the hours go. The 90-minute proposal flips this entirely: you spend your creative energy on the problem statement and let AI handle the heavy lifting on everything else.
"A proposal that used to take a full day now takes 90 minutes — and it reads better than what most salespeople produce when they're grinding through it manually."
Before You Open Claude: The Discovery Notes
The quality of your AI-generated proposal is directly proportional to the quality of your discovery notes. Before you write anything, pull together the following from your calls and research:
- The prospect's stated problem in their own words
- The measurable impact of that problem (time lost, revenue missed, customers churned)
- Their desired outcome — what does success look like to them specifically?
- Any objections or concerns they raised
- Key stakeholders and their priorities
- Timeline and urgency signals
If you can't fill out that list from memory after your discovery call, you need to go back and ask better questions before writing anything.
The 90-Minute Workflow
Minutes 0–15: Feed Claude Your Discovery Notes
Open Claude and give it everything you collected above. Ask it to write the problem statement section of your proposal — two to three paragraphs that articulate the prospect's situation, the impact of the problem, and what's at stake if they don't address it. Review it, adjust any details, and make sure it sounds like it came from your discovery conversation — because it should.
Minutes 15–35: Build the Solution Section
Tell Claude your standard service offering and ask it to frame it specifically around this prospect's situation. Not generic service descriptions — a narrative that connects your solution directly to the problem you just articulated. Claude will give you a strong first draft. Your job is to review and add any specifics it couldn't know from your notes.
Minutes 35–50: Investment and ROI Framing
Drop in your pricing and ask Claude to help frame the investment in terms of ROI — connecting the cost to the impact of the problem you described. A $5,000 engagement looks very different when it's framed against $40,000 in annual waste or 200 hours of recovered time. This section alone improves close rates when done well.
Minutes 50–70: Why Us and Social Proof
Ask Claude to write the credibility section using the case studies, results, and testimonials you provide. Give it the raw material — the numbers, the client names (if approved to use), the specific outcomes — and let it shape them into a compelling narrative. Then review and tighten.
Minutes 70–90: Final Review and Personalization Pass
Read the entire proposal as if you're the prospect. Does every section connect back to what they actually said in discovery? Does it feel like it was written for them specifically, or does it feel like a template with their name swapped in? Make the human edits that AI can't — the specific reference to the comment they made, the acknowledgment of their timeline, the detail that shows you were listening.
What This Isn't
This workflow doesn't produce generic proposals. It produces personalized proposals faster. The personalization still comes from you — from your discovery, your judgment, your understanding of what this specific prospect needs to see to say yes. AI handles the structure, the framing, and the first draft. You handle the insight and the final judgment call.
Never send an AI-generated proposal without a full review. Claude doesn't know things you didn't tell it. Check every specific claim, every number, every reference to the prospect's situation. Your credibility is on the line — not Claude's.
The Compounding Benefit
The first time you run this workflow it might take two hours. The tenth time it takes 75 minutes. The fiftieth time you're at 60 minutes with a win rate that's gone up because your proposals are more consistently personalized than they were when you were writing everything from scratch.
That's the compounding benefit of a documented AI workflow. It gets faster and better over time without you working harder.
Take the Free AI Blueprint Quiz →