How Stanton Ridge Raised Its Proposal Win Rate from 31% to 44%
A boutique strategy firm cut proposal time from 16 hours to 5 and raised win rate from 31% to 44%. The variable wasn't the AI tool — it was what they fed it.
Stanton Ridge Advisory, a twelve-consultant boutique strategy firm, spent sixteen hours building each proposal.
Six months after implementing an AI-assisted proposal system: five hours.
Win rate: 31% to 44%.
Estimated additional year-one revenue: £200,000.
Those numbers come from Brightbots, an AI automation firm specializing in management consulting, which documented the Stanton Ridge case in an April 2026 analysis. The outcome data is Brightbots' own reported client results — and the finding buried in the methodology matters more than the headline numbers.
The dependency that made it work
Brightbots was direct about what produced the win rate improvement: it wasn't switching proposal tools. It was what Stanton Ridge brought to the proposal tools.
"Spend time tagging past proposals by sector, engagement type, and outcome so the AI agent can retrieve the most relevant material." — Brightbots, "AI Automation for Management Consulting Firms," April 17, 2026
The upstream work — sorting, tagging, and organizing the firm's archive of past proposals — was the investment that produced the output quality. The proposal AI was doing retrieval and synthesis. What it retrieved and synthesized was only as good as what it could find.
Rubbish in, rubbish out.
Why the tool isn't the differentiator
AI proposal tools have reached commodity status across boutique advisory. According to srl-sasame.com's 2026 analysis of proposal automation in consulting, when all advisors use the same tools, proposals converge — and that's the deeper problem the Stanton Ridge result exposes. The differentiator is what makes the outputs distinct: the uniqueness of the engagement data feeding them.
An advisor who applies AI proposal generation to a well-organized repository of past proposals — tagged by sector, engagement type, win/loss outcome, and client context — gets outputs that reflect the firm's actual knowledge and history. An advisor who applies the same tools to a raw collection of files and email threads gets something more generic.
When proposals start sounding alike, the engagement data layer is where differentiation lives.
What this means for a practice at Stanton Ridge's scale
Stanton Ridge didn't achieve the win rate improvement because they found a better tool. They achieved it because they did the upstream work first: organizing their engagement history in a format that gave the AI something specific to work with.
The standard boutique proposal process — per Brightbots, 12 to 20 hours, starting from a template or whatever you can pull together from the last similar engagement — produces a proposal that reflects whoever's writing it right now, with whatever they can recall or quickly retrieve. Brightbots puts industry-average win rates at 30% to 35%. That's the baseline.
The upgrade isn't a better tool. It's a better input layer: past proposals organized by type, outcome, and sector; client intelligence that persists between engagements; win themes that can be surfaced and applied. When that layer exists, AI proposal generation produces outputs that reflect the firm's accumulated knowledge. When it doesn't, the AI produces outputs that reflect the template and whatever you could assemble in the last hour before the proposal was due.
Stanton Ridge's 44% win rate sits 9 to 14 percentage points above the industry average. That gap isn't a function of the AI tool. It's a function of what the tool had to work with.
The managed alternative
Building and maintaining an organized engagement intelligence layer requires two things most boutique practices struggle to sustain: initial curation effort and ongoing maintenance discipline. The initial work is a project most practices can start. Ongoing maintenance is where it breaks down — curation competes directly with billable hours, and the marginal cost of keeping the system current never goes away.
The managed alternative separates the curation problem from the benefit. The practice gets the organized intelligence layer. The overhead of maintaining it lives elsewhere.
That's the structure Fieldway Intelligence Services (FIS) is built around — a maintained engagement intelligence layer that makes past proposals, client context, and engagement history retrievable in real time, not reconstructed from memory before each new proposal.
The proposal AI tools are table stakes. The win rate improvement comes from what you give them.
If you want to see what the organized engagement data layer looks like for a practice at your stage, email matthew@fieldway.org.
Related: Knowledge Debt: The AI Prerequisite Your Practice Hasn't Solved Yet | The 30-Client Plateau Is a Knowledge Problem, Not a Headcount Problem | The $50K Alternative to McKinsey: Managed Intelligence for Advisors
More from Intelligence Services
The Consulting Firm AI Margin Gap Has Nothing to Do With Tools
The 2026 SPI benchmark shows a 17.9% vs. 6.0% EBITDA gap. The difference isn't which AI tools consulting firms use — it's what's organized underneath them.
The 30-Client Plateau Is a Knowledge Problem, Not a Headcount Problem
When boutique advisory practices hit the capacity ceiling, most reach for headcount. What we see consistently: it's a knowledge architecture problem, not a bandwidth problem.
The Fractional COO's Role Just Changed. Here's the Gap It Left.
AI drove a 46% fractional COO demand surge by restructuring the role itself. Here's what the unbundling left behind – and why it matters for boutique advisors.
Drowning in client documents?
Fieldway Intelligence Services is a managed document intelligence offering for boutique advisory firms — we ingest, synthesize, and maintain the deliverables that matter, so you stay in your zone of genius.
Talk to me about FIS