The other night I built a 90-day revenue projection tool with my Codex executive assistant.
The tool lets me look at projected income and revenue for the business over the next 90 days.
But the important part is not just that AI built a tool.
The important part is why the tool is useful. It has context.
Codex is already helping me operate the business:
- it helps with transactions
- it updates KPI workflows
- it understands my Notion structure
- it sees the numbers I track
- it knows the operating rhythm of the business
So when I ask it to build a projection tool, it is not starting from a blank prompt. It is building from real context.
Then I take the tool and show it to my executive chat inside ChatGPT.
That chat gives me feedback on how to make the tool better.
So now the loop looks like this:
Business data -> Codex builds the tool -> ChatGPT critiques the tool -> Codex improves the tool -> I make better decisions.
That is the part most people are missing with AI.
They are still using it like a one-off answer machine.
But the leverage is when AI has enough context to help you see the business more clearly.
At some point, you have to stop guessing. Look at the numbers. Build the tool. Improve the tool. Make the decision cleaner.
I attached screenshots of the tool so you get an idea what it looks like.
Quick question: What is one decision in your business that you are still making mostly off feel instead of data?