Introduction
AI is making waves, but you need to see results worth adopting team-wide. What's the catch? The AI model's responses can be hit-or-miss, even when the queries are very similar in nature. How can you trust AI when it seems so inconsistent?
The hiccup in the grand scheme of things is the quality of your AI inputs.
It's a common challenge among occasional users of generative AI—you've been told over and over again that the key to success lies in being specific in your instructions, but it's hard to anticipate everything the AI might misinterpret. What level of specificity should you aim for? How can you make sure AI's writing stays clear and on-topic the whole way through?
You only need a simple prompt formula that easily integrates into your everyday tasks.
Use the RISEN framework to generate content that exceeds expectations
Credit to Kyle Balmer, who created this framework and shares lots of other great stuff on prompt engineering!
The RISEN framework breaks down the process of crafting effective prompts into five key parts: Role, Instructions, Steps, End Goal, and Narrowing. With all these elements strategically placed in your prompt, you'll talk to the AI model in a way that's clear and focused.
And for all those leading teams, it's worth mentioning RISEN doesn't just make AI more accessible; it makes your team more proactive and innovative. When everyone has the tools and the know-how to leverage AI effectively, they start seeing opportunities for improvement everywhere they look.
As a leader, that's exactly what you want: a team always looking for ways to optimize processes, save time, and deliver better results.
Role: Define the AI's persona
The RISEN framework's first step is defining the AI's role or persona. Assigning a clear role to the AI sets the stage for a more focused, productive interaction. This targeted approach already minimizes the chances of receiving irrelevant or off-topic responses.
When you hire a new team member, you always provide them with a clear job description that outlines their responsibilities and the skills required for the role. The same principle applies to AI. By defining the AI's role, you give it specific skills and knowledge to work with.
Look at the difference between a prompt that goes straight into the task, "Create a Transitioning Team Member Responsibilities During Absence SOP," and one that starts with:
"As a detail-oriented team leader..."
The latter phrase establishes the AI's role, priming it to generate content with the expertise and nuances of a diligent team leader.
Instructions: Give clear and precise directives
Once you've defined the AI's role, the next step is to provide clear, precise instructions. The key here—and where most people struggle—is to articulate exactly what you need the AI to do, leaving no room for ambiguity or misinterpretation.
At the time of publishing, Similarweb data shows people usually spend around 8 minutes and 15 seconds on ChatGPT during each visit. This might be longer than you'd expect, and it's likely due to the fact that so many keep ChatGPT on hand and spend more time trying out different ways of asking questions to get the best possible answers.
AI is smart, but like humans, it can't read your mind either! It might focus on things that aren't relevant to your target audience. Let's continue our SOP example:
"As a detail-oriented team leader, create a comprehensive SOP that outlines the steps for transitioning responsibilities from one team member to another in the event of an extended absence."
Without clear instructions, you might create a document that is too long, bloated with fluff, or lacks tactical steps for a team member to follow.
This applies to other tasks, such as writing an email response, organizing tables in a report, or creating social media copy for different channels. The more specific and detailed your instructions, the better the AI can understand and meet your needs with fewer revisions.
Now we're warmed up and ready to give the AI model steps to follow!
Want to see this in action? Join the ClickUp Showcase, a weekly AMA-style webinar where you can participate as little or as much as you want in a group demo setting.
Steps: Guide AI through the process
After giving clear instructions, the RISEN framework encourages you to guide the AI through the process by outlining specific steps or components you want to include in the final output.
This is where you become the architect, breaking down complex tasks so the generated content flows logically from one point to the next.
Stay with our SOP example:
"As a detail-oriented team leader, create a comprehensive SOP that outlines the steps for transitioning responsibilities from one team member to another in the event of an extended absence. Begin by categorizing the tasks into sections such as communication protocols, project handover, ongoing responsibilities, and contingency plans."
Your guidance will significantly enhance the quality and uniformity of the AI's output, making it easier for your target audience to follow along and engage with it.
Here's a side-by-side comparison of two different prompts (and we're not even done with the RISEN framework). The left image looks like a project brief, whereas the right image goes straight into the SOP table of contents.
End Goal: Clarify the purpose
The fourth component of the RISEN framework is clarifying your request's end goal or purpose. This is where you communicate the bigger picture and help the AI model understand the broader strategy of the task.
Here's the SOP prompt that tells the AI model you don't just want tasks and due dates; you want a tool that empowers new team members to execute a smooth transfer of knowledge and responsibilities:
"As a detail-oriented team leader, create a comprehensive SOP that outlines the steps for transitioning responsibilities from one team member to another in the event of an extended absence. Begin by categorizing the tasks into sections such as communication protocols, project handover, ongoing responsibilities, and contingency plans. Include clear descriptions and timelines for each task, and consider adding links to relevant resources or documentation. Ensure the SOP is well-organized, easy to follow, and helps the team maintain continuity and productivity during the transition period."
Ideally, the end goal should be placed toward the end of your prompt after you've provided the AI with the necessary context, instructions, and steps.
This placement allows the AI to process all the relevant information before focusing on the ultimate objective.
Narrowing: Set boundaries for focused creativity
The final component of the RISEN framework involves setting boundaries and constraints to focus the AI's creativity. You won't restrict the AI model's ability to generate unique, original content with parameters. In fact, providing specific parameters can actually lead to more effective results.
Jeff Su, a Product Marketer at Google, takes it a step further by providing the AI model with a KPI to not only save him time from reworking the output but also avoid irrelevant suggestions.
Whether it's word count limitations, formatting preferences, or stylistic guidelines, these constraints will help you harness the power of AI while still maintaining control over the final output.
You could add a few formatting guidelines to the SOP, such as:
"Present the information using numbered lists, bullet points, and bolded section headers."
Or...
"Include a table of contents at the beginning of the document to provide an overview of the SOP's structure and make it easier for users to find specific sections."
Generate content that informs, inspires, and engages
By breaking down the process into five manageable components—Role, Instructions, Steps, End Goal, and Narrowing—RISEN debunks AI communication and helps you craft specific, time-saving, and purposeful prompts.
But why stop there?
When you combine the RISEN framework with a tool like ClickUp Brain, the potential for productivity and efficiency is limitless. ClickUp Brain's conversational, contextual, and role-based AI features and ability to leverage your organization's collective knowledge make it the perfect companion for your AI-powered workflow.
Unlike other AI models that require separate browsers, context, and constant switching between applications, ClickUp Brain is accessible directly within your Workspace to help with tasks, docs, and more.
See it in action at the ClickUp Showcase and ask questions to our experts to learn more about integrating AI into daily workflows!
