Library › Leadership AI › Consultants

Plan week one of an engagement as a hypothesis-driven table

  • For Consultants
  • You practice Setting limits and rules
  • Time about 5 minutes
  • Level Advanced

About this exercise

Plan the first five working days of the pricing engagement as a markdown table with columns Day, Hypothesis to test, Data or interview, Owner. Exactly 5 rows, Mon to Fri. Each hypothesis is a statement that could be proven wrong. Only use data or interviews that the notes say are available on that day, and do not schedule anything on a day the source is unavailable. Output only the table.

Common mistakes when you ask AI for this

  • Using unavailable data. The AI schedules the competitor price list. Try: Say: "Only data available that day."
  • Scheduling on the wrong day. The AI puts the CFO on Wednesday. Try: Say: "Respect availability: CFO Fri only."
  • Questions, not hypotheses. The AI writes 'Why did discounts rise?'. Try: Say: "Each hypothesis is a statement that could be proven wrong."
  • Wrong shape. The AI adds columns or rows. Try: Say: "Columns: Day, Hypothesis to test, Data or interview, Owner; five rows."
  • Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands over it. Try: Cap each cell: "each cell under 14 words."
  • Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands over it. Try: Cap each part instead, for example: "each line under 12 words."
  • Adding headings and bold. The AI wraps the table in headings, bold goals and notes. Try: Say: "Plain text: only the table, no headings or bold."

What a strong prompt includes

  • Columns
  • Five rows
  • Testable
  • Availability
  • Owner

Practice this now ↑

Related exercises