Skip to main content

Chapter 3: Common Mistakes to Avoid

Learn From Others' Errors​

Even after learning good prompting techniques, there are common pitfalls that trip people up. Knowing these mistakes—and how to fix them—will save you frustration.

Mistake 1: Being Too Vague​

❌ The Problem​

Write something about marketing.

AI has no idea: What format? What length? What aspect of marketing? For whom?

✅ The Fix​

Write a 200-word LinkedIn post about content marketing strategies 
for B2B companies. Focus on practical tips for small marketing teams. 
Professional but conversational tone.

Before/After Example​

BeforeAfter
"Help me with my resume""Review my resume summary below and suggest 3 specific improvements. I'm applying for senior software developer roles. Focus on highlighting leadership experience."
"Explain AI""Explain what machine learning is in 3 sentences for someone who's never heard of it. Use an everyday example."
"Write a story""Write a 300-word mystery story set in a small-town bakery. Include a surprising twist at the end."

Mistake 2: Overloading with Instructions​

❌ The Problem​

Write me a blog post about productivity, make it engaging but also 
informative, use statistics but not too many, include examples but 
keep it concise, make it suitable for beginners but don't 
oversimplify, add a call to action but make it subtle, optimize for 
SEO but don't keyword stuff, use subheadings and lists and images 
and quotes and...

Too many conflicting requirements = confused output.

✅ The Fix​

Focus on the most important requirements:

Write a 500-word blog post about morning productivity routines.
Audience: busy professionals
Format: intro, 3 tips with subheadings, conclusion
Tone: practical and encouraging
Include one statistic and one real example.

The Rule​

5-7 key requirements is usually the sweet spot. If you have more, consider breaking into multiple prompts.

Mistake 3: Not Iterating​

❌ The Problem​

First response isn't perfect → User gives up or starts over completely.

✅ The Fix​

Treat the first response as a starting point, not the final answer.

Iteration Examples​

First response too long?

Shorten this to 100 words, keeping the key message.

Wrong tone?

Rewrite in a more casual, friendly tone. Less corporate-speak.

Missing something?

Good, but add a specific example for each point.

Not quite right?

This is close. Make these changes:
- Remove the second paragraph
- Add more detail in the conclusion
- Make the headline catchier

Mistake 4: Expecting Perfection​

❌ The Problem​

  • Thinking AI should read your mind
  • Expecting human-quality judgment without guidance
  • Assuming the first response is as good as it gets

✅ The Fix​

  • AI is a tool, not a mind reader
  • Guide and refine for better results
  • Edit the output - add your voice, correct errors, personalize
  • Verify facts - AI can make mistakes or hallucinate information

Reality Check​

AI output is a first draft, not a finished product. Your job is to:

  1. Generate a starting point
  2. Review critically
  3. Refine and improve
  4. Add your personal touch
  5. Verify any facts or claims

Mistake 5: Common Pitfalls and Fixes​

Pitfall: Using Relative Terms Without Anchors​

❌ Wrong:

Make it shorter.
Write something brief.
Keep it simple.

✅ Right:

Make it 100 words or less.
Write 3-4 sentences.
Use vocabulary a 10-year-old would understand.

Pitfall: Assuming AI Knows Your Context​

❌ Wrong:

Write about our product launch.

AI: What product? What company? What industry?

✅ Right:

Write about the launch of our new mobile app "TaskMaster" - 
a productivity app for remote workers. Our company is a 
startup called FocusTech targeting small business owners.

Pitfall: Asking Multiple Unrelated Things​

❌ Wrong:

Write me an email to my boss, also create a workout routine, 
and explain quantum physics.

✅ Right: Ask each separately. Each task gets focused attention and better results.

Pitfall: Giving Up After One Bad Response​

❌ Wrong:

[gets bad response]
"AI doesn't work for this"
[gives up]

✅ Right:

[gets bad response]
"That wasn't quite right. Let me clarify: [more specific prompt]"
[iterates until good]

Pitfall: Not Providing Examples When Needed​

❌ Wrong:

Write in my company's voice.

AI has no idea what your company sounds like.

✅ Right:

Write in my company's voice. Here's an example of how we write:
"[paste example of your actual content]"
Match this tone and style.

Before/After: Complete Transformations​

Example 1: Email Request​

Before:

Write an email for me.

After:

Write a brief professional email to a client named Alex who submitted a late invoice. I need to remind them politely but firmly about our 30-day payment terms. Maintain the relationship while being clear about expectations. Maximum 100 words.

Example 2: Content Creation​

Before:

Blog post about fitness.

After:

Write a 400-word blog post titled "3 Simple Exercises You Can Do At Your Desk." Target audience: office workers who sit all day and have back pain. For each exercise, explain: how to do it (3 steps), why it helps, how often to do it. Friendly, encouraging tone. End with motivation to start today.

Example 3: Problem Solving​

Before:

Help me with my business.

After:

I run a small online t-shirt shop. Sales have dropped 30% over the last 3 months. My main channels are Instagram and Etsy. I've been posting less frequently lately. Help me diagnose possible causes and suggest 3 specific actions I can take this week to start turning things around.

🎯 Practical Exercise: Fix These Prompts​

Improve each of these prompts:

Prompt 1: "Write something for my website"

  • What would you add?

Prompt 2: "Explain everything about investing"

  • How would you narrow this?

Prompt 3: "Make my resume better"

  • What specifics are needed?

Try writing improved versions, then test them with AI.

Quick Reference: The Fix Checklist​

Before sending a prompt, check:

  • ☐ Specific enough? Did I specify format, length, audience?
  • ☐ Focused? Am I asking for one thing, not five?
  • ☐ Context provided? Does AI know my situation?
  • ☐ Clear terms? Are my requirements concrete, not vague?
  • ☐ Ready to iterate? Do I have follow-ups planned?

The Master Formula​

When in doubt, use this structure:

[TASK] Write/Create/Explain/Analyze [what you need]

[FORMAT] In the form of [format] that is [length]

[AUDIENCE] For [who will use this]

[CONTEXT] Background: [relevant situation]

[SPECIFICS] Include: [must-haves]
            Avoid: [what not to include]
            Tone: [how it should sound]

Quick Summary​

  • Vague prompts → Add specifics (format, length, audience)
  • Overloaded prompts → Focus on 5-7 key requirements
  • First response not perfect → Iterate and refine
  • Expecting perfection → View AI output as a first draft
  • Use the fix checklist before sending prompts

Course Complete! 🎉​

Congratulations! You've completed all three modules of Introduction to Prompt Engineering.

What You've Learned:​

  • ✅ What prompts are and why they matter
  • ✅ The iteration mindset
  • ✅ Being clear and specific
  • ✅ Providing useful context
  • ✅ Using examples (few-shot learning)
  • ✅ Practical writing prompts
  • ✅ Problem-solving with AI
  • ✅ Avoiding common mistakes

What's Next?​

  1. Take the Module 3 Quiz
  2. Review the Glossary for key terms
  3. Take the Final Exam
  4. Explore more Resources
  5. Practice! The best way to improve is to use these techniques daily

You're now equipped to get 10x better results from AI. Go put these skills to work!