Purpose
Use this lesson to make missing inputs visible before drafting. The aim is to produce work that a colleague can review and act on, not merely text that sounds convincing.
Lesson Explanation
Scenario: Create a sales proposal for a new prospect with almost no details.
Better instruction: Before drafting, ask up to five questions about the customer problem, scope, budget range, timeline, and decision maker.
The fastest route is sometimes a question, not an answer. Tell the model to pause when a missing fact would materially change cost, commitment, audience, or advice. Limit the questions so discovery does not become a questionnaire. The questions address problem, scope, budget, timeline, and decision maker.
Work in a short cycle: provide the relevant input, state the result and limits, inspect the draft against the acceptance criteria, then revise the instruction when a requirement is missing. Keep the final human decision with the person responsible for the work.
Practice Questions
-
A prompt asks an AI to “create a marketing plan” for an unnamed product with an unspecified budget and no target audience. What does this lesson say should happen here?
View Answer
The AI should pause and ask clarifying questions, since missing facts like budget, audience, and product details would materially change the plan.
-
What does this lesson say is “sometimes the fastest route,” instead of an immediate answer?
View Answer
A question.
-
A prompt for a legal document is missing the specific jurisdiction it needs to comply with. Should the AI attempt to guess and produce a draft anyway, per this lesson?
View Answer
No – a missing fact that would materially change the advice (like jurisdiction) should prompt a clarifying question rather than a guess.
-
What five areas does this lesson give as an example of what a clarifying-questions instruction might cover for a sales proposal?
View Answer
The customer problem, scope, budget range, timeline, and decision maker.
-
A prompt tells the AI to ask up to five clarifying questions before drafting. Why does this lesson specify a limit on the number of questions?
View Answer
To keep discovery from becoming an overwhelming questionnaire – the limit keeps the clarification process efficient.
-
A team asks an AI to “write our hiring policy” without specifying which country’s employment law applies, a fact that would materially affect the content. What does this lesson recommend?
View Answer
The AI should ask about this missing fact before drafting, since it would materially change the content.
-
A prompt for a minor formatting change (e.g., “make this shorter”) is missing a specific target word count. Does this lesson suggest this always warrants a full round of clarifying questions?
View Answer
Not necessarily – the lesson’s principle is about facts that would materially change cost, commitment, audience, or advice; a minor stylistic gap may not always rise to that threshold.
-
Why does this lesson caution against letting “discovery become a questionnaire”?
View Answer
Excessive questioning can slow down the process and become as unhelpful as guessing – the goal is a limited, targeted set of genuinely necessary clarifying questions.
-
A prompt for a customer proposal is missing the prospect’s budget range, a fact that would significantly change what’s proposed. What does this lesson recommend the AI do?
View Answer
Ask about the budget range before drafting, since it materially affects cost and commitment.
-
A prompt is designed to have the AI ask about scope, timeline, and decision maker before drafting a project proposal. What is the purpose of asking about “decision maker” specifically, per this lesson’s framing?
View Answer
It reflects a category of missing fact that would materially change how the proposal should be written or targeted.
-
What kinds of missing facts does this lesson say should trigger a clarifying question, rather than a guess?
View Answer
Facts that would materially change cost, commitment, audience, or advice.
-
A request for an event budget is missing the expected number of attendees, which would significantly change the numbers throughout. Based on this lesson, what should the AI do instead of estimating a number on its own?
View Answer
Ask for the missing attendee count, since it’s a fact that would materially change the resulting cost figures.