The Art of Prompting: Getting Better AI Outputs (Topic 1) in Module 2 – AI-at-Work (BG)

The Art of Prompting: Getting Better AI Outputs

Why Prompting Matters

The same AI model produces wildly different quality outputs depending on how you ask. Effective prompting is the skill that separates professionals who get measurable productivity gains from AI and those who give up after a few frustrating attempts.

The Four Elements of a Strong Prompt

1. Role — Tell the model who it is:

"You are an experienced marketing strategist with expertise in SaaS B2B companies."

2. Task — Be specific about what you want:

"Write a 5-bullet competitive analysis comparing our product to Competitor X, focusing on pricing and integration capabilities."

3. Context — Give the model what it needs to know:

"Our product is [X], priced at [Y], targeting [Z customers]. Competitor X is [description]."

4. Format — Specify how you want the output:

"Use a two-column table. Keep each bullet under 25 words. Write at an executive audience level."

Prompt Patterns That Work

Chain of thought: "Think through this step by step" — improves accuracy on reasoning tasks.

Few-shot examples: "Here are two examples of what I'm looking for: [example 1] [example 2]. Now do the same for: [your task]."

Persona assignment: "Act as a skeptical CFO reviewing this business case. Identify the three weakest financial assumptions."

Constrained output: "Answer in 3 bullet points maximum. No introductory text. No conclusions."

Iteration: Don't expect perfection from the first prompt. Use follow-ups: "Make this more concise", "Add a risk section", "Rewrite the second point to be less aggressive".

What Prompting Cannot Fix

  • Factual hallucinations on topics outside the training data
  • Real-time data requirements
  • Tasks requiring genuine domain expertise the model wasn't trained on
  • Judgment calls that require organizational context
Sign in to join the discussion.
Recent posts
No posts yet.

The Art of Prompting: Getting Better AI Outputs

Why Prompting Matters

The same AI model produces wildly different quality outputs depending on how you ask. Effective prompting is the skill that separates professionals who get measurable productivity gains from AI and those who give up after a few frustrating attempts.

The Four Elements of a Strong Prompt

1. Role — Tell the model who it is:

"You are an experienced marketing strategist with expertise in SaaS B2B companies."

2. Task — Be specific about what you want:

"Write a 5-bullet competitive analysis comparing our product to Competitor X, focusing on pricing and integration capabilities."

3. Context — Give the model what it needs to know:

"Our product is [X], priced at [Y], targeting [Z customers]. Competitor X is [description]."

4. Format — Specify how you want the output:

"Use a two-column table. Keep each bullet under 25 words. Write at an executive audience level."

Prompt Patterns That Work

Chain of thought: "Think through this step by step" — improves accuracy on reasoning tasks.

Few-shot examples: "Here are two examples of what I'm looking for: [example 1] [example 2]. Now do the same for: [your task]."

Persona assignment: "Act as a skeptical CFO reviewing this business case. Identify the three weakest financial assumptions."

Constrained output: "Answer in 3 bullet points maximum. No introductory text. No conclusions."

Iteration: Don't expect perfection from the first prompt. Use follow-ups: "Make this more concise", "Add a risk section", "Rewrite the second point to be less aggressive".

What Prompting Cannot Fix

  • Factual hallucinations on topics outside the training data
  • Real-time data requirements
  • Tasks requiring genuine domain expertise the model wasn't trained on
  • Judgment calls that require organizational context
Sign in to join the discussion.
Recent posts
No posts yet.
Info
You aren't logged in. Please Log In or Join for Free to unlock full access.