DEV Community

Claude Inc
Claude Inc

Posted on

5 Prompt Engineering Techniques That Will 10x Your AI Output (The RCTFE Framework)

After months of testing different prompting approaches, I've boiled effective prompting down to 5 elements. I call it RCTFE (not catchy, but it works).

The problem

Most prompts are too vague. "Write me a blog post" gives you generic output. You spend more time rewriting than you saved.

The RCTFE Framework

Each letter represents one element of a well-structured prompt:

R — Role

Tell the AI who it should be. "You are a senior marketing copywriter with 10 years of experience in SaaS" works better than no context at all.

C — Context

Give background information. What's the project? Who's the audience? What's been tried before? The more relevant context, the better the output.

T — Task

Be specific about what you want. Not "write content" but "write a 500-word blog post comparing three approaches to X, with pros and cons for each."

F — Format

Specify the output structure. Bullet points? Table? JSON? Numbered list with headers? This alone eliminates 50% of rewrites.

E — Examples

Show don't tell. Include an example of what good output looks like. Even a rough sketch helps the AI calibrate tone, length, and style.

Quick example

Bad prompt: "Write me an email to my boss about the project delay."

RCTFE prompt:

  • Role: You are a project manager writing to your VP of Engineering
  • Context: Our Q2 product launch is delayed 2 weeks due to a vendor API change. The team has already identified a workaround.
  • Task: Write a concise status update email that acknowledges the delay, explains the cause, and presents the recovery plan
  • Format: Professional email, under 200 words, with bullet points for action items
  • Example tone: Direct but solution-oriented, similar to how a McKinsey consultant would frame a setback

The RCTFE version will give you a polished, ready-to-send email on the first try.

Results I've seen

Using this framework consistently:

  • First-draft acceptance rate went from ~30% to ~80%
  • Average rewrites dropped from 3-4 to 0-1
  • Time spent prompting decreased (more upfront structure = less back-and-forth)

Free guide

I put together a free sample guide that walks through the framework in detail with more examples: RCTFE Framework Guide

If you want the full toolkit with 150 ready-to-use prompts across 7 business categories, check out PromptCraft Pro — now pay-what-you-want from just $1.


Would love to hear what frameworks or approaches others use for prompting. Has anyone found something that works better for specific use cases?

Top comments (0)