AI content approval dropped 60% → 26% in 2 years. The D.E.P.T.H Method fixed it.

Anyone else getting called out for janky AI-sounding writing? Discover how to write effective AI prompts that produce authentic, engaging, and high-quality AI generated content.

The Data Is Brutal:

Consumer enthusiasm for AI content plummeted from 60% in 2023 to a paltry 26% in 2025.

People can spot generic, AI-generated writing easily now. This highlights the importance of prompt engineering to help AI systems produce better results.

The phrases that set off those "AI Detector" alarm bells:

• That tired "Let's delve into…"
• "It's important to note…"
• Cliché phrases like "In today's fast-paced world…"
• And of course "Unlock the power of…"

Here's What's Going On:

MIT researchers found that vague prompts cause AI tools to go haywire and produce generic, unhelpful content because the AI system can't get a clear picture of what we want.

Most users write prompts like:

• Write a blog post about AI marketing
• Create a LinkedIn post about productivity

The result? Vague input = generic AI produced output. Every. Single. Time.

The Solution: The DEPTH Method for Writing Better Prompts

After testing over 1000 + AI prompts, this formula consistently beats simple prompts and eliminates that awkward, robotic tone:

D – Define Multiple Perspectives

Wrong: "You're a marketing expert"
Right: "Imagine you're three experts working together: a behavioural psychologist figuring out decision triggers, a conversion copywriter crafting persuasive language, and a data analyst looking at performance metrics"

Why it works: It forces the AI model out of single-perspective "default generic mode" and into multi-dimensional thinking, stimulating creativity and improving the model's focus.

E – Establish Clear Success Metrics

Wrong: "Make it good"
Right: "Must achieve: conversational tone (grade 8 reading level), exactly one clear Call To Action, under 150 words, optimized for 40%+ open rate, and avoid clichéd phrases like 'delve into'"

Why it works: Clear instructions help AI systems understand exactly what "good" means, leading to better AI generated content.

P – Provide Context Layers

Wrong: "For my business"
Right: "Context: B2B SaaS, $200/mo product, target audience: burnt-out founders aged 35-50, previous campaign emails averaged 20% opens (goal: 35%+), industry: productivity tools, brand voice: direct but empathetic, competitor analysis: [give me some examples]"

Why it works: Providing more context helps AI produce tailored and accurate responses, reducing generic guessing.

T – Task Breakdown

Wrong: "Write the whole email"
Right:

1. What's the #1 pain point this audience is feeling?
2. Come up with a pattern-interrupt hook that doesn't use clichés
3. Build some credibility with specific data/examples
4. Add a soft CTA with a clear next step

Why it works: Breaking down the task into smaller parts prevents AI systems from jumping straight into generic templates and improves output quality.

H – Human Feedback Loop (The Game Changer)

Wrong: Accepting the first output
Right: "Rate this output 1-10 on: originality (no AI clichés), clarity, persuasion power. Flag any generic phrases. If anything scores below 8, revise it. Compare to top-performing emails in [industry] and see where we're missing out."

Why it works: Self-critique catches "AI slop" before publishing, ensuring the AI tool produces engaging and authentic written content.

Real Impact:

The Billion Dollar Boy research found that audiences aren't rejecting AI, they're rejecting BAD AI.

When we use structured prompting and prompt engineering:

• AI stops relying on generic templates
• Output matches our unique voice
• Content passes the "sounds human" test

The Time Investment:

Yes, DEPTH takes 5 minutes vs. 30 seconds for "write a blog post."

But would you rather:

• 30 seconds + 30 minutes editing generic output = 30.5 minutes
• 5 minutes upfront + minimal editing = 8 minutes total

Want the Exact Prompts?

I've spent months testing and documenting 1,000+ AI prompts using DEPTH across every scenario (emails, social posts, blog content, sales copy, technical docs). Each prompt includes:

• The complete DEPTH structure
• Success metrics defined
• Context templates
• Self-critique loops
• Before/after examples

Check my full collection. It'll save you 6+ months of trial-and-error in writing prompts.

The Bottom Line:

AI isn't getting worse, our prompts are just falling behind what audiences now expect. DEPTH closes that gap and helps AI produce better results.

What's your experience?