What Is Prompt Engineering?

Prompt engineering is the practice of writing clear, structured instructions that help ChatGPT produce the response you actually want. ChatGPT generates its output based entirely on what you give it — the more precise and intentional your input, the more useful and accurate the result.

You do not need any technical background to write effective prompts. The principles are straightforward: be specific, provide context, and tell ChatGPT what kind of output you expect. This article walks through practical techniques you can apply immediately.

The Fundamentals

Be Specific

The single most impactful thing you can do is be specific about what you want. Vague prompts produce vague results. Compare:

Vague: "Tell me about enrollment."

Specific: "Summarize fall 2025 enrollment trends at UCCS, broken down by college. Focus on year-over-year changes and highlight any colleges with enrollment growth or decline greater than 5%."

The second prompt tells ChatGPT the scope (fall 2025, UCCS), the structure (broken down by college), the focus (year-over-year changes), and the threshold that matters (5%). It will produce a far more useful response.

Assign a Role

Telling ChatGPT who it should act as helps frame the tone, depth, and perspective of the response. This is especially useful when you need domain-specific expertise or a particular communication style.

Examples:

  • "You are an experienced academic advisor. A first-generation college student asks you how to choose a major. Walk them through your thought process."
  • "You are a technical writer. Rewrite the following error message so that a non-technical end user can understand it."
  • "You are a grant reviewer for the NSF. Evaluate the following abstract and identify its strengths and weaknesses."

Define the Output Format

If you have a specific format in mind, say so. ChatGPT will match the structure you describe.

Examples:

  • "Respond in a numbered list."
  • "Format your response as a table with columns for Name, Department, and Status."
  • "Write this as a professional email with a subject line."
  • "Give me a one-paragraph summary followed by three bullet points with key takeaways."
  • "Respond in plain language at an 8th-grade reading level."

Without format instructions, ChatGPT will choose its own structure — which may or may not be what you need.

Provide Context

ChatGPT has no access to your internal systems, institutional knowledge, or the specifics of your situation unless you tell it. The more relevant context you provide, the better the output.

Instead of: "Write a response to this complaint."

Try: "I work in the Registrar's Office at UCCS. A student emailed complaining that their transfer credits were not applied to their degree audit. Our policy is that transfer credit evaluation takes 4–6 weeks after official transcripts are received. Write a professional, empathetic response that acknowledges the student's frustration, explains the timeline, and provides the contact information for our Transfer Credit team at [email protected]."

This gives ChatGPT the role, the situation, the policy, the tone, and the specific details to include.

Intermediate Techniques

Break Complex Tasks into Steps

For multi-part tasks, break your request into sequential steps rather than asking for everything at once. This gives ChatGPT a clear path to follow and reduces the chance of it skipping something important.

Instead of: "Create a presentation about our new employee onboarding process."

Try: "I need to build a presentation about our new employee onboarding process. Let's work through this step by step:

  1. First, suggest an outline with 8–10 slide titles that cover the full onboarding journey from offer acceptance through the 90-day check-in.
  2. Once I approve the outline, write the content for each slide.
  3. After that, suggest talking points I can use as speaker notes."

This approach also gives you checkpoints to review and redirect before ChatGPT goes too far down a wrong path.

Use Examples

If you want ChatGPT to match a specific style, tone, or format, give it an example to follow. This is one of the most effective techniques available.

Example: "Here is an example of how we write knowledge base article summaries:

'Multi-Factor Authentication (MFA) adds a second layer of security to your UCCS account. After entering your password, you will be prompted to verify your identity using the Microsoft Authenticator app, a text message, or a phone call. MFA is required for all faculty, staff, and students.'

Now write a similar summary for our new VPN service."

ChatGPT will match the length, tone, and structure of your example closely.

Specify What to Include and Exclude

Telling ChatGPT what not to do is just as important as telling it what to do.

Examples:

  • "Explain how DNS works. Do not use analogies — just describe the technical process directly."
  • "Summarize this research paper. Focus only on the methodology and findings. Do not include the literature review or future work sections."
  • "Write a job description for a systems administrator. Do not include salary information."
  • "List the pros and cons of migrating to a cloud-hosted LMS. Do not recommend a specific product."

Set Constraints

Constraints help keep responses focused and appropriately scoped.

Examples:

  • "Explain this concept in 3 sentences or less."
  • "Limit your response to 200 words."
  • "Provide exactly 5 recommendations."
  • "Use only information from the uploaded document — do not use your general knowledge."

Ask ChatGPT to Think Step by Step

For tasks that involve reasoning, analysis, or problem-solving, explicitly asking ChatGPT to work through the problem step by step can significantly improve the quality of the output.

Example: "A student has 45 transfer credits from a community college and needs 120 credits to graduate. They want to complete their degree in 2 years taking a full course load. Walk me through the math step by step to determine if this is feasible and what course load they would need per semester."

This technique — sometimes called chain-of-thought prompting — encourages ChatGPT to show its reasoning rather than jumping to a conclusion, which often produces more accurate and transparent results.

Advanced Techniques

Iterate and Refine

You do not need to get the perfect prompt on the first try. Treat ChatGPT as a collaborative tool — start with a reasonable prompt, evaluate the output, and then refine.

Useful follow-up prompts:

  • "This is good, but make the tone more formal."
  • "Shorten this to half the length."
  • "Add a section about accessibility requirements."
  • "Rewrite the third paragraph to focus more on cost savings."
  • "That's too technical — simplify it for a non-technical audience."
  • "Try again, but this time structure it as a FAQ."

Iteration is not a sign of a bad prompt — it is a normal and effective part of the workflow.

Use System-Level Instructions in Custom GPTs and Projects

If you find yourself repeating the same instructions across many conversations, consider embedding those instructions in a Custom GPT or a Project.

For example, if you frequently write KB articles, you could create a Custom GPT with instructions like: "You are a technical writer for UCCS Identity Services. Write all content in plain language, use active voice, and follow the UCCS KB formatting standard. Always include a 'Getting Started' section at the end of each article."

Every conversation with that Custom GPT will follow these instructions automatically, eliminating the need to re-specify them each time.

Provide Reference Material

When ChatGPT has access to your actual documents, it can produce much more accurate and grounded responses. Upload relevant files — policy documents, style guides, data files, or examples of previous work — and instruct ChatGPT to reference them.

Example: "I've uploaded our department's style guide and a previous annual report. Using the style guide's formatting standards and the tone from last year's report, draft the executive summary section for this year's report based on the data I'll provide next."

Ask for Multiple Options

When you are exploring ideas or need to choose between approaches, ask ChatGPT to generate several options rather than committing to one.

Examples:

  • "Give me three different subject lines for this email, ranging from formal to conversational."
  • "Suggest five possible titles for a workshop on data privacy in higher education."
  • "Write two versions of this announcement — one for a student audience and one for faculty."

This gives you options to compare and helps you clarify what you actually want.

Use ChatGPT to Improve Your Prompts

If you are not sure how to phrase a request, you can ask ChatGPT for help.

Examples:

  • "I want to ask you to help me write a grant proposal, but I'm not sure what information you need from me. What questions should I answer before we start?"
  • "Here's my prompt: [paste prompt]. How could I improve this to get a better response?"
  • "What additional context would help you give me a more useful answer about our network architecture?"

This is a surprisingly effective technique — ChatGPT can often identify exactly what information is missing from your request.

Common Mistakes to Avoid

Being too vague. "Help me with a report" gives ChatGPT almost nothing to work with. Specify the topic, audience, format, length, and purpose.

Asking for too much at once. A prompt that asks ChatGPT to research a topic, draft a 10-page report, create an executive summary, and suggest presentation slides will produce mediocre results across the board. Break it into stages.

Accepting the first response without review. ChatGPT can produce confident-sounding output that contains errors, omits important details, or misunderstands your intent. Always read critically and iterate.

Not providing enough context. ChatGPT does not know your institution's policies, your team's preferences, or the specifics of your project unless you provide that information. When responses feel generic, the fix is usually more context.

Assuming ChatGPT remembers previous conversations. Each new conversation starts fresh (unless you are using Projects or Memory). If context from a previous chat is important, re-provide it or work within a Project that retains it.

Over-relying on a single prompt pattern. Different tasks call for different approaches. A prompt that works well for drafting emails may not work well for data analysis or creative brainstorming. Adapt your technique to the task.

Quick Reference: Prompt Checklist

Before sending a complex prompt, run through this checklist:

  • [ ] Role: Did I tell ChatGPT who it should act as?
  • [ ] Task: Is the task clearly stated?
  • [ ] Context: Did I provide the relevant background information?
  • [ ] Format: Did I specify how I want the response structured?
  • [ ] Constraints: Did I set boundaries on length, scope, or content?
  • [ ] Examples: Would an example help ChatGPT match my expectations?
  • [ ] Exclusions: Did I mention anything ChatGPT should avoid?

Not every prompt needs all of these — a quick question does not require a role assignment and format specification. But for important or complex tasks, covering these elements will consistently produce better results.

Further Resources

For more comprehensive guidance on prompting techniques, OpenAI provides detailed documentation at https://platform.openai.com/docs/guides/prompt-engineering.

For questions about ChatGPT Edu at UCCS, see the related articles in this knowledge base.

  • No labels