Purpose
Use this lesson to request a structure that can be pasted, reviewed, or imported. The aim is to produce work that a colleague can review and act on, not merely text that sounds convincing.
Lesson Explanation
Scenario: Compare three vendors for a procurement meeting.
Better instruction: Return a Markdown table with Vendor, annual cost, integration risk, support hours, and open questions. Do not invent missing values.
Formatting is a functional requirement. Specify headings, column names, field order, and what to do with missing information. A well-shaped output reduces manual cleanup and makes review faster. The table has the five named columns and marks unknown values instead of guessing.
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 “list our top five risks” with no specification of format. What does this lesson say about formatting requirements?
View Answer
Formatting is a functional requirement – headings, column names, field order, and handling of missing information should be specified.
-
A request for a comparison of two software tools doesn’t specify column names or structure. What is the likely result, based on this lesson?
View Answer
An unstructured or inconsistently formatted response, requiring more manual cleanup before it can be used.
-
Rewrite “List our top five risks” into a prompt with a specific, usable output format.
View Answer
Something like: “Return a table with columns: Risk, Likelihood (High/Medium/Low), Impact, and Suggested Mitigation. List the top five risks only.”
-
What does this lesson say a well-shaped output reduces?
View Answer
Manual cleanup, and it makes review faster.
-
A prompt asks for missing information to be marked clearly rather than guessed at. What principle does this reflect?
View Answer
Specifying what to do with missing information, one of the functional formatting requirements this lesson identifies.
-
A prompt for a budget comparison doesn’t specify what should happen if a cost figure is unknown, and the AI fills it in with an estimate presented as fact. What could have prevented this?
View Answer
Specifying in the prompt that unknown values should be marked as unknown rather than invented, per this lesson’s guidance on formatting.
-
A request for a project status table specifies column names but not their order. Is this a complete output-format specification, per this lesson?
View Answer
Not necessarily complete – field order can also matter for usability, and this lesson lists it as one of the elements to specify.
-
Why does this lesson treat output format as “a functional requirement” rather than a stylistic preference?
View Answer
Because the shape of the output directly affects whether it can be used efficiently – a well-shaped result reduces manual cleanup and speeds up review.
-
A prompt specifies “return a Markdown table” but doesn’t name the columns. What element from this lesson’s guidance is still missing?
View Answer
Column names (and potentially field order and missing-value handling).
-
A team receives an AI-generated list of options with no structure at all – just a paragraph of running text. What might this lesson suggest was missing from the original prompt?
View Answer
A specified output format, such as a table, numbered list, or defined structure, rather than leaving the shape of the response undefined.
-
A prompt for a competitor comparison specifies headings, column names, and field order, and instructs the AI not to invent missing values. What does this reflect?
View Answer
A fully specified, usable output format, consistent with all the elements this lesson identifies.
-
Why might a beautifully written but unstructured paragraph fail to meet the “usable output format” standard this lesson describes, even if factually accurate?
View Answer
Because usability depends on structure that reduces manual cleanup and speeds review, not just accuracy – an unstructured paragraph may still require significant rework to use.