Proposal Win Strategy AI Companion

read

Twennie’s Proposal Win Strategy AI Companion

How to Teach Your AI to Conduct a Candid, Evidence-Based Proposal Review

AUTHOR

Michele Rochon-Wood

Artificial intelligence has become remarkably good at generating proposal content. What many professionals have not yet discovered is that AI can be even more valuable as a proposal reviewer.

Unfortunately, a generic instruction such as review this proposal usually produces a generic response. The AI may comment on sentence length, grammar, passive voice, or formatting without examining whether the proposal presents a strong competitive argument.

A Training Manual for Your AI

The Proposal Win Strategy AI Companion is not another chatbot or AI platform. It is an editable instruction set that teaches ChatGPT, Microsoft Copilot, Claude, Gemini, Fuel iX, and similar systems how to evaluate a technical proposal using Twennie’s Ten Steps to a Win Theme methodology.

Think of the Companion as an onboarding manual for a new proposal manager. You would not hand a newly hired employee a proposal and simply ask what they thought. You would explain your review process, describe what matters to the client, and teach them how to recognize strong evidence and genuine differentiation.

The AI Companion provides that same kind of training to artificial intelligence. Instead of writing a new prompt for every proposal, you can create a dedicated proposal evaluation assistant and teach it the methodology once.

Ten Sections of Strategic Evaluation

The Companion instructs the AI to evaluate the proposal in ten separate sections. Each section examines one part of the proposal’s competitive argument.

The AI evaluates whether the proposal understands the client, understands the project, recognizes the issues, identifies critical success factors, and addresses important gaps in the proposal team’s knowledge.

It then evaluates the features of the firm’s offer, the benefits those features create, the evidence that the promised benefits have been delivered before, and whether that evidence truly differentiates the firm from its competitors.

Finally, it determines whether all of those elements have been woven together into one clear proposal story.

Build a Dedicated Proposal Evaluation Assistant

The accompanying video shows you how to create a reusable proposal evaluation assistant using the AI platform available to you.

In ChatGPT, you can create a Custom GPT and upload the Companion as a permanent knowledge document. In Claude, you can create a Project that retains the Companion between reviews. Microsoft Copilot users can build an assistant within the Microsoft environment, while Google Workspace users can use Gemini with documents stored in Google Drive.

Larger firms may use enterprise AI systems such as Fuel iX, which provide controlled access to several AI models through one secure organizational environment. The exact setup differs by platform, but the central process remains the same: create a dedicated assistant, provide the evaluation framework, and use that assistant consistently.

Give the AI Enough Information

AI can only evaluate the information available to it. Uploading the draft proposal alone may produce useful observations, but it will not allow the system to determine whether important client knowledge, project information, or competitive evidence was omitted.

For a stronger review, provide the Request for Proposal, client meeting notes, site visit observations, pursuit strategy documents, proposal planning outputs, project profiles, lessons learned, proposal debriefs, and competitor intelligence whenever those materials are available.

The Companion includes a section identifying which supporting documents are most valuable for each of its ten evaluation stages. These recommendations also provide proposal teams with a useful checklist for gathering and managing pursuit intelligence.

Receive a Candid Blue Team Review

Once the Companion, proposal, and supporting documents have been uploaded, the AI conducts a structured Blue Team review.

For every section, it provides a score, identifies evidence found in the proposal, highlights missing evidence, explains the competitive risks created by weaknesses, and recommends practical improvements. It also assigns a confidence level so the proposal team can see whether its conclusions are based on strong evidence or limited information.

The Companion explicitly instructs the AI not to reassure the team that the proposal is good enough. Its job is to conduct a candid assessment of the submission’s chances in a competitive marketplace and identify improvements while there is still time to make them.

Think Before You Rewrite

One of the most important instructions in the Companion appears on its final page: do not rewrite the proposal immediately.

Proposal teams should review the evaluation together, discuss the findings, decide which recommendations are valid, and agree on the strategy before asking AI to produce replacement text.

AI should function first as a reviewer and strategic advisor. Only after the proposal team has exercised its professional judgment should the system be asked to strengthen an executive summary, revise a section, improve a win theme, or suggest better headings.

Edit the Companion for Your Firm

The Proposal Win Strategy AI Companion is provided as an editable template. Firms can adapt the instructions, terminology, scoring system, and evaluation standards to reflect their own proposal process.

Over time, the assistant can also be given additional organizational knowledge, including proposal standards, writing guidelines, project databases, debrief records, lessons learned, and established review criteria.

This allows a firm to build a proposal evaluator that applies the same standards across offices, teams, and pursuits while still supporting the judgment of experienced proposal professionals.

The First Twennie AI Companion

The Proposal Win Strategy Evaluator is the first in a planned collection of Twennie AI Companions. Future resources may help AI evaluate proposal readability, project risk, scope management, business development strategy, client relationships, leadership, communication, and other areas of technical consulting work.

Eventually, these methodologies will become part of Twennie’s own integrated AI. Instead of teaching another AI platform how to conduct the evaluation, members will be able to upload their proposals and supporting documents directly into Twennie.

The methodology will remain the same. Only the delivery platform will change.

Twennie’s goal is not to replace proposal managers or professional judgment. It is to give technical professionals access to a thoughtful, disciplined, and consistent reviewer before their work reaches a client.