PracticeCritical Thinking & First Principles
A plausible number, unsourced
The benchmark the model supplied
The situation
Your developer-tools product added a free tier four months ago. Hana has drafted the annual plan: a revenue model and a hiring plan for four people. The model's central assumption is a free-to-paid conversion rate the team asked an AI assistant for. The board pack goes out on Friday.
Hana · Finance lead: “The board pack goes out Friday. It’s a standard range; do you want me to change anything?”
The plan is well built and the range is plausible. The free tier's dashboard is open next to it. You have time to check one thing before Friday.
What you can see (4) · tap a card to read it
›E01 · The planning premise
The plan states an 'industry-standard 3–5% free-to-paid conversion' and uses a 3.5% midpoint. The figure is attributed to an AI assistant's answer, quoted in a footnote. No other source is given.
›E02 · What rests on it
At 3.5% the model projects enough new paid accounts to fund four hires starting next quarter: two engineers, one support, one sales. At 1.5% the same model funds one hire. The hiring plan is written for four.
›E03 · The document
Twelve pages, clearly structured, with a sensitivity table, three scenarios and hedged language throughout. It reads like the work of a careful analyst.
›E07 · The dashboard headline
Free tier, four months: 2,400 free accounts, 29 paid. The dashboard divides one by the other and shows 1.2%.
Step 1 of 2
What do you check?
Your first choice is kept. Changing your mind later counts as a retry.