The reality is simple: the quality of an AI's output is often directly related to the quality of the prompt it receives. Learning how to communicate effectively with AI can dramatically improve the relevance, accuracy, and usefulness of the results.
In our recent System Skills Recharge training session, MBS Sales Consultant Heather Gillespie led a session on how to use practical AI to elevate workplace skills (the full webinar is available to MBS system customers in the Resource Center). Today we will go over a simple framework for AI prompts and seven best practices.
A Simple Framework for Better Prompts
When creating prompts, remember this formula:
Role + Goal + Context + Audience + Requirements + Output Format
Here's an example:
Act as a campus store manager. Analyze the attached textbook adoption report for the upcoming term. The audience is academic department chairs and college administrators. Explain how adoption submission rates compare to prior terms, identify the top three trends affecting textbook affordability and availability, and highlight any risks associated with late adoptions. Write in a professional and data-driven tone. Limit the response to 250 words. Format the output as an executive summary followed by three key findings, identified risks, and recommended actions.
This framework provides the AI with everything it needs to deliver focused, relevant, and actionable results.
The better you communicate your objectives, context, and expectations, the better the output will be. The most successful AI users aren't necessarily the most technical. They're the ones who have learned how to ask better questions.
7 Best Practices That Help You Get the Most From AI Tools
1. Define the Goal
Every successful prompt starts with a clear objective. Before you begin typing, ask yourself: What do I want the AI to accomplish?
Vague instructions such as "Write something about faculty adoptions" leave too much room for interpretation. Instead, be specific about the task, desired outcome, tone, style, length, and any constraints.
For example: Create a concise email that explains why faculty benefit from submitting on-time adoptions and the impact timely textbook adoptions have on students. Highlight the top three findings and include recommended actions.
Placing the primary instruction at the beginning of your prompt helps the AI prioritize what matters most.
2. Provide Context
Context engineering is one of the most important ingredients in effective prompting. AI tools perform best when they understand the background, business environment, audience expectations, and any relevant data. The more relevant information you provide, the more tailored and accurate the response is likely to be.
Consider including:
- Background information
- Relevant facts and data
- Project objectives
- Constraints and requirements
- Previous examples or reference materials
3. Assign a Role
AI can adapt its approach based on the role you assign it. By specifying who the AI should act as, you can influence both the expertise and communication style used in the response.
Examples include:
- Act as a marketing strategist.
- Write as a communications consultant.
- Act as an experienced bookstore manager.
When you assign a role, the AI is better able to tailor its recommendations, vocabulary, and perspective to the task at hand. For example, you could ask AI to create a template email from an experienced bookstore manager to campus stakeholders that highlights a recent sales report. This would give you text to use as a foundation when sharing bookstore information with administrators without having to share actual bookstore data with AI.
4. Specify the Audience
One of the most common mistakes users make is forgetting to identify the audience.
A message intended for senior administrators should look very different from one written for students. The audience influences the language, level of detail, tone, and structure of the content.
Be clear about who will read the final output. For example: Write a concise website description for this t-shirt, no more than 25 words. The primary audiences are students, alumni and fans. (include an image of the t-shirt in the request)
This simple instruction can significantly improve the effectiveness of the response.
5. Set Clear Requirements
Think of AI as a highly capable assistant that benefits from detailed instructions.
If you want the AI to explain its thought process, compare options, prioritize recommendations, or follow a particular methodology, tell it explicitly.
Examples include:
- Explain your reasoning step-by-step.
- Provide three recommendations ranked by impact.
- Include risks, benefits, and next steps.
- Compare the options and recommend the best solution.
Establishing clear requirements reduces ambiguity and increases the likelihood of receiving a useful response on the first attempt.
6. Be Clear and Specific
Specificity is the secret ingredient of great prompting.
Avoid unclear or conflicting instructions that can confuse AI. Clearly define what success looks like and specify the format you want.
Common output formats include:
- Emails
- Website copy
- Presentation outlines
- Posters
The more specific the format and expectations, the easier it is for the AI to deliver exactly what you need.
7. Review and Refine
Even the best prompts rarely produce a perfect result on the first attempt. Instead of viewing the initial response as the final answer, treat it as a draft.
Refinement is a critical part of the AI workflow. Try follow-up instructions such as:
- Make this more concise.
- Rewrite in a more professional tone.
- Add more detail to the recommendations.
- Simplify this for a general audience.
Prompting is an iterative process. Small adjustments can often lead to significant improvements in output quality.
Upcoming System Skills Recharge Training
Beyond Discounts: Driving Sales Without Sacrificing Margin
Date: August 13, 2026
Time: 12:00–1:00 p.m. CT
In this session, you’ll:
- Develop an understanding of retail markdown theory
- Understand how pricing decisions impact sales, inventory turns, and gross margin
- Utilize MBS system tools to create and manage pricing actions that support inventory and sales goals
- Develop pricing strategies to move slow selling or seasonal merchandise while minimizing margin erosion
- Evaluate sales and inventory data to make more informed decisions about when and how to adjust pricing
No advanced tech skills required. MBS system customers can register now to attend.
