I’ve been publicly exploring AI’s role in sustainability for nearly a year now — sharing real use cases, tools, and workflows. And one comment keeps popping up: “AI isn’t good enough for this work.” I get it. The tech is evolving fast, and there’s real concern about job security. But dismissing AI outright isn’t the answer. As sustainability professionals, our job is to explore, adapt, and see what’s possible.
Today’s use case? Solving one of the most painful moments in Scope 3 onboarding.
Scope 3 explanations rarely go smoothly — 15 categories, scattered data owners, and just enough ESG jargon to lose your audience in slide two.
How did I try to fix that? Let’s dive in.
Scope 3 is complicated — and it’s quickly becoming a hot KPI across industries. Everyone wants to know how you're measuring value chain emissions, but actually calculating them? That’s where it gets messy. As a sustainability manager, you’re expected to report on 15 distinct categories — using data that sits with HR, finance, procurement, and other teams far outside your domain. Which means those teams need to understand what’s required and why it matters. And that’s the real challenge.
The Spark: What if AI Could Make It Visual and Interactive?
The Scope 3 challenge is massive — but the best way to tackle a big problem is one small, practical step at a time. I asked myself: what’s the first friction point we can actually fix?
For me, it was this: during onboarding calls or team briefings, it’s incredibly hard for people to visualize the full Scope 3 landscape. As a finance or procurement lead, it’s not clear what your role is, where you fit, or what data you're responsible for.
So I focused on solving just that.
I wanted a simple, dynamic dashboard — something that displays all 15 categories in one place, allows you to click through for definitions, methods, and data owners, and makes it easy to jump back and forth during a live discussion.
That alone would instantly make onboarding smoother. But even better — it could be the foundation for something much bigger.
Stick around until the end, and I’ll show you how this small start opens the door to powerful extensions.
The Build: From Idea to App in 10 Minutes
Once I had the concept in mind, I didn’t overthink it — I just wanted to see if I could build something functional and fast. No slides. No dev brief. Just ChatGPT, Replit, and a clear idea.
a. Preparing the Inputs
To get started, I gathered three core inputs:
My client’s Scope 3 onboarding table, which included the 15 categories, description of methods, relevant data owners, and emission factors
The GHG Protocol Corporate Value Chain (Scope 3) Standard, to ensure technical alignment
A brand color palette — not essential, but since I already had one ready, I figured: why not make it look polished?
b. The Tools
ChatGPT acted as my product owner. In typical software teams, the product owner bridges the gap between what users need and what developers build. I used ChatGPT to do just that — translating sustainability context into a build-ready product brief.
Replit was the developer — a kind of magician that turns plain English into working code. I didn’t write or tweak any code myself.
c. The Workflow
The build process came down to two major steps:
Step 1: GPT Prompt Preparation
I fed ChatGPT everything: tables, references, goals, and features. The better you explain your idea — with clear context and detail — the better the output.
I use ChatGPT for three main reasons:
It structures and simplifies the prompt for Replit to understand
It creates a reusable conversation thread — useful whether you expand the app later or port it to another platform
It reduces cost — fewer iterations inside Replit means fewer compute credits spent
Step 2: Replit Magic
I then copied the final prompt(s) from ChatGPT into Replit. It instantly generated a code preview with a short description of what the app would do. Once I confirmed everything looked right, I hit “Run.”
Within 5–10 minutes, the interactive app was live.
The Reveal: The Scope 3 Onboarding Tool
The result? A simple, interactive dashboard that makes Scope 3 onboarding smoother and more visual. You can click through all 15 categories, see descriptions, data needs, and who’s responsible — all in one place.
🎥 Watch the full walkthrough here
The Toolkit: Replicate It Yourself
Want to build your own version? I’ve packaged everything into a Notion page you can copy and customize.
🧰 The Scope 3 Onboarding Toolkit includes:
A structured table with all 15 categories, methods, and data owners
The original ChatGPT prompt template I used