AI won’t replace you.
BUT people who know how to use it will.
Everyone has access now. Everyone has accounts. Everyone is “experimenting.”
The real gap is no longer tools. It’s how you use them.
Now comes the harder part:
learning how to use it properly.
Here are the 7 deadly prompting sins… and exactly how to fix them.
Sin 1: No context
You write:
“Analyse this.”
The model guesses your role, your audience, and your constraints.
Guessing multiplies errors.
Fix:
Add role + task scope + constraints.
Not more words.
Clearer ones.
Example:
“You are a product analyst. Analyse the attached transcript for founders from seed to Series A to find the vision outliers. Output a 5-bullet decision memo. Max 180 words.”
Sin 2. Vague Instructions
You write:
“Write about marketing trends.”
Trends for who?
For when?
For what decision?
You didn’t define success, so the output can’t hit it.
Fix:
Define acceptance criteria.
Audience, timeframe, format, proof points.
Clarity is leverage.
Example:
“Write a 1,000-word brief on the three most important B2B AI marketing trends for Q3 2025. Include one data point per trend with a source and a one-line implication.”
Sin 3. Treating AI like Google
You write:
“What are good onboarding ideas?”
This gets you surface-level fluff.
Questions get answers.
Directives get deliverables.
Fix:
Turn questions into jobs.
Flows. Steps. KPIs. Tables.
Tell it what to produce.
Example:
“Draft a 5-step onboarding flow for a B2B SaaS. Include the email subjects, timing in days, and one KPI per step.”
Sin 4. Asking for everything at once
One giant prompt feels efficient.
It hides failure and creates spaghetti output.
Different tasks require different thinking modes.
Fix:
Split work into stages.
Chain outputs intentionally.
AI performs best when you guide the sequence.
Example:
Step 1: “List the 5 core customer jobs-to-be-done with a one-line pain for each.”
Step 2: “Using the chosen JTBD 2 and 4, write homepage H1 options (5) within 8 words each.”
Step 3: “Expand H1 #3 into a 150-word hero section.”
Sin 5. Not iterating
You ask once.
You judge forever.
But this is a conversation, not a vending machine.
Fix:
Critique → revise → narrow.
Angles first.
Then titles.
Then structure.
Then draft.
Iteration is the multiplier.
Example:
Step 1: “List 5 potential angles for my article about [topic] to get eyeballs from [audience]. Focus on [niche].”
Step 2: “Using the chosen angle, let’s write 10 SEO-ready titles with their one-liners summaries.”
Step 3: “Using the chosen titles and one-liner, let’s create the outline of the article to…”
Sin 6. No format or tone
No constraints = generic voice.
If you don’t force shape,
the model defaults to bland.
Fix:
Specify structure and tone.
Length. Sections. Voice. Banned words.
Good prompts sound opinionated.
Example:
“Write a LinkedIn post. 220 words. Hook (2 lines), 3 bullets, one CTA. Tone: direct and practical, no buzzwords, plain English.”
Sin 7. No examples
Rules help.
Examples teach taste.
Fix:
Add 1–2 gold standards.
Optionally one anti-example.
You’re training preferences, not intelligence.
Example:
“Model the tone and density on these two snippets [paste]. Avoid this anti-example [paste]. Keep sentence length under 16 words.”
The real shift
Stop asking AI to “help.”
Start assigning it work.
In 2026, the advantage won’t be using AI.
It’ll be directing it better than everyone else.
Prompting isn’t a trick.
It’s a skill and skills compound.
If you guilty of any of the 7 sins, contact us to help improve your prompting skills.