Write efficient prompts using the CREATE formula (applicable to all AI).

This guide provides detailed instructions on how to write AI prompts using the CREATE formula to improve accuracy and output control.

The use of AI tools like ChatGPT, Gemini, or Claude is becoming increasingly popular, but not everyone knows how to properly 'command' the AI. For the same question, someone who knows how to write prompts might receive a clear, relevant answer, while someone else might receive a generic or irrelevant response.

That's why the CREATE formula was created — a framework that helps you build logical, complete, and more controllable prompts. CREATE consists of five components: Character, Request, Example, Adjustment, and Type of Output. When combined correctly, you can almost 'program' how the AI ​​thinks and responds.

C – Character (Defining the role of AI)

The first step is to 'assign a role' to the AI. Instead of having the AI ​​give generic answers, you ask it to play the role of a specific expert.

This helps the AI ​​select the right tone, knowledge, and perspective to suit the context.

For example, instead of asking:

Write a marketing plan

You can write:

 Act as an experienced digital marketing manager. Write a marketing plan

Adding roles helps make the answers more realistic and insightful.

R – Request (Specific Request)

This is the most important part: you need to clearly state what you want.

A common mistake is writing requests that are too vague. AI isn't as good at 'guessing' what you mean as you think, so the more specific you are, the more accurate the results will be.

For example:

Write about AI

Compared to:

Explain how AI is used in e-commerce to increase sales

The second version is clearer, so the results are more useful.

E – Example (Provide an illustrative example)

AI works very well when it has examples to 'mimic'. If you provide one or two examples, the AI ​​will understand the format and presentation style you want.

This is especially useful when writing content, emails, or formatting data.

For example:

Here is an example: Product: Coffee Maker Description: Compact coffee maker suitable for small kitchens

Now write a similar description for a blender

Having examples helps to make the output much more consistent and formatted.

A – Adjustment (Adjustment, Constraint)

This section allows you to control the output using specific conditions such as length, tone, target audience, or what to avoid.

Without Adjustment, AI often responds in a 'general' way—sometimes not tailored to the objective.

For example:

Keep the tone friendly and simple, avoid technical jargon, under 150 words

Or:

Make it suitable for beginners

These constraints help AI 'narrow its scope' and produce results that are more closely aligned with needs.

T – Type of Output

You need to specify what format you want the result in: paragraph, bullet points, table, checklist, etc.

Unless otherwise specified, the AI ​​will choose the default format — which may not be what you want.

For example:

Provide the answer in bullet points

Or:

Create a comparison table

By simply adding this one line, the output can be completely transformed, becoming much easier to read and use.

A complete prompt example using the CREATE formula.

When you combine all these elements, you'll have a much clearer and more powerful prompt.

Act as an experienced UX designer. Explain how to improve a mobile app onboarding flow. Here is an example of good onboarding: simple steps, clear instructions, minimal input. Keep it beginner-friendly and under 200 words. Provide the answer in bullet points.

Compared to the simple prompt, this version helps the AI ​​understand better:

• Who are you playing?
• What needs to be done?
• What format?
• What are the limits?

The results, therefore, are much closer to the actual needs.

When should you use CREATE?

CREATE is particularly useful in cases including:

• Write content (blog posts, emails, social media posts)
• Analyze or explain the problem.
• Create structured documents
• Working with complex data or workflows

For simple prompts, you might not need all five steps. But when the request becomes complex, CREATE helps you avoid repeatedly prompting and re-prompting.

CREATE isn't a rigid formula, but rather a way of thinking when working with AI. Instead of writing prompts intuitively, you can define roles clearly, describe specific requirements, and provide examples if needed. Then add constraints and choose the right format. With consistent application, you'll see a significant improvement in the quality of AI responses—fewer revisions, fewer follow-up questions, and significantly more time savings.

You've just finished reading the article "Write efficient prompts using the CREATE formula (applicable to all AI)." edited by the TipsMake team. We hope this article has provided you with many useful tech tips and tricks. You can search for similar articles on tips and guides. Thank you for reading and for following us regularly.