Mindvalley Stories
Case Study · AI Mastery

A sales professional mapped the RFP-to-SOW workflow and built three custom GPTs

Three custom GPTs form a copilot for SaaS professional services sellers, built on a workflow where 60–70% of manual effort proved pattern-based.

This student looked hard at one of the most tedious stretches of selling professional services, turning a request for proposal into a statement of work, and broke it into small, automatable steps. From that map came three custom GPTs designed to work together as a copilot. It is a clear example of what happens when someone stops wishing their workflow were easier and starts taking it apart.

Khanh Nguyen
A builder working on professional services selling for SaaS companies, using ChatGPT, Claude and prompt engineering.

Copilot framework for Professional Services Sellers

Before

The work of moving from an RFP to a signed SOW was a long chain of manual steps. Professional services sellers at SaaS companies carried that effort by hand, from reading requests to shaping proposals and scopes. Much of it felt like it had to be done fresh every time.

“Decomposed the full RFP-to-SOW workflow into granular, automatable steps — identifying that 60–70% of manual effort is pattern-based and suitable for AI automation”

In their own words

The turning point

During the program, they decomposed the full RFP-to-SOW workflow into granular steps. Looking at each one closely, they found that 60–70% of the manual effort was pattern-based and suitable for AI automation. That realisation turned a vague idea into a concrete design.

After

They designed and built all three custom GPTs that form the Copilot framework for professional services sellers. Working with ChatGPT, Claude, prompt engineering and Google Workspace, they turned their workflow map into tools meant to work side by side. Their submission calls them an Automation Architect, and the work fits the name.

What it lets people do

The copilot is designed to take on the repeatable, pattern-based parts of the RFP-to-SOW process. That leaves sellers more room for the judgement and relationships that patterns cannot cover.

What they learned

Before automating anything, you have to see the work clearly, step by step. Breaking a process down is what revealed how much of it could be handed to AI, and where people still matter most.

Copilot framework for Professional Services Sellers was built during AI Mastery.

Read the full submission
1) Achievement Designed and built all three Custom GPTs that will form the Copilot framework for Professional Services Seller for SaaS companies 2) Tools: ChatGPT, Claude, Prompt Engineering, Google Workspace 3) Methodology Decomposed the full RFP-to-SOW workflow into granular, automatable steps — identifying that 60–70% of manual effort is pattern-based and suitable for AI automation 4) Learnings Please find details in the [Automation Architect] AI_Mastery_Awards_Submission document 5) Relevant URLs: Please find details in the [Automation Architect] AI_Mastery_Awards_Submission document file:///Users/pknguyen2025/Downloads/Final%20Version%20copilot-visual-light.html

Written 14 February 2026 Submitted by the storyteller

Every story is written by a Mindvalley student or by someone our teachers have worked with. How we collect and check stories

More from AI Mastery

Copilot framework for Professional Services Sellers: a AI Mastery case study | Mindvalley Stories