ChatGPT is already better than most consultants, coaches, and managers - we're f***** ...
Theory of Constraints EBITDA Throughput Accounting
Simplified example for priorities based on Throughput Accounting // constraint usage
ChatGPT is already better than most consultants, coaches, and managers - we're f***** ...
March 19, 2025
I did an experiment yesterday - and I thought that this must be a simple Turing test. I never thought that AI was already advanced enough to get it right without hints!
99% of managers fail at this question, and most coaches do too. So I asked the AI (ChatGPT o3-mini-high) - and oh shit...
...it solved the problem in 9 seconds and could even explain how!
What is it about?
In a company or team, you always have fewer resources than ideas - so you have to prioritize.
This is not so easy - neither ROI nor WSJF nor other methods provide an overview of everything, and so this leads time and again to blatantly wrong decisions - usually, ‘then we just do everything’.
But there is a solution that, although not perfect, finds a 99% solution for optimal prioritization with minimal effort!
The concept is called the "Theory of Constraints" by Eliyahu Goldratt!
Do you know it?
No, it's stupid because ChatGPT is one step ahead of you!
Here is the question - plain and simple
A company consists of three teams
Team A - 10 people available for projects per month
Team B - 10 people available for projects per month
Team C - 5 people available for projects per month
Every month, a person can work 10 days on projects
The company deals with two classes of projects
Project Class 1 needs 10 days from Team A and, if finished, 50 days from Team C
Project Class 2 needs 50 days from Team B and, if finished, 25 days from Team C
A project of Class 1 promises sales of 60.000$
A project of Class 2 promises sales of 50.000$
Which project class should sales focus on to get the best bottom-line results for the company?
Which class of project should sales focus on?
Before you read further, try to solve it on your own!
p.s.: ChatGPT even answered how much more profit the company gets - without being asked!
The result (ChatGPT o3-mini-high) ...
... Let's break down the problem step by step:
Monthly Team Capacities
- Team A: 10 people × 10 days = 100 days available
- Team B: 10 people × 10 days = 100 days available
- Team C: 5 people × 10 days = 50 days available
Project Requirements
- Project Class 1: Requires 10 days from Team A. Requires 50 days from Team C. Sales: $60,000 per project
- Project Class 2: Requires 50 days from Team B. Requires 25 days from Team C. Sales: $50,000 per project
Identifying the Bottleneck Resources
- For Class 1: Team A: With 100 available days, you could handle up to 100/10 = 10 projects. Team C: With 50 available days, you can only handle 50/50 = 1 project. Bottleneck: Team C limits you to 1 project per month, generating $60,000 in sales.
- For Class 2: Team B: With 100 available days, you could handle up to 100/50 = 2 projects. Team C: With 50 available days, you can handle 50/25 = 2 projects. Bottleneck: Both teams allow for 2 projects per month, generating 2 × $50,000 = $100,000 in sales.
Conclusion
Given the limited capacity of Team C (50 days per month), Project Class 2 makes more efficient use of that resource. By focusing on Class 2 projects, the company can complete 2 projects per month (as opposed to just 1 for Class 1), resulting in higher monthly sales ($100,000 vs. $60,000).
Sales should therefore focus on Project Class 2 to achieve the best bottom-line results for the company.
Perfect! Sorry, I was stunned. Most managers don't see the constraint and don't know about prioritization by the octane number = throughput/effort in the constraint.
Just out of interest, I asked ...
What is the name of the management concept used for this reasoning, and who invented it?
The reasoning is based on the Theory of Constraints (TOC), which focuses on identifying and addressing the bottleneck - the resource that limits overall performance. This concept was introduced by Eliyahu M. Goldratt in his book The Goal.
By identifying that Team C was the bottleneck resource, the analysis directed focus toward the project class that made the most efficient use of that limited capacity, thereby maximizing the company's output and bottom-line results.
What do you think?
What do other LLMs with reasoning do - can they also answer?
Want to learn more and stay at least one step ahead of the AIs? We founded a community of humans to learn more about hyper-productivity - open and free:
Join the biggest network of TOC and Self-Organization experts for free
p.s.: Maybe we should invite an AI to our community too ... no - we stay human