Jason Pikoos Enhancing AI Chatbot Interactions with Context-Rich Personas
Imagine having access to the expertise of a lawyer, accountant, or warehouse manager, wherever and whenever you like, for pennies. I’ll show you how.
Last month’s issue laid the foundation for effectively prompting Generative AI-powered chatbots (such as ChatGPT, Claude, and Gemini). This month, I’ll delve into customizing these interactions to suit your business needs better through the use of personas.
What is a persona?
Think of a persona as a character the chatbot adopts during interactions. A persona encompasses tone, expertise, language style, personality traits, and response tendencies. The clearer we define these elements, the better the AI can tailor its responses to meet our specific needs. For instance, you can expect a more useful response to a request to resolve your laptop issue if you first instruct the chatbot to be a “laptop support specialist”.
Building a persona
Here’s a summary of the key elements to consider when creating an effective persona. While it’s not necessary to include every detail, generally, the more information you provide, the better and more tailored the response will be.
- Job Title and Experience
Specify the desired job title, position, and years of experience that you want. You can also consider any relevant professional qualifications. You can never make the chatbot too overqualified.
Example: “You are a Human Resources Director with a SHRM-CP qualification and 20 years of experience in handling employee relations, tracking HR metrics, and managing compensation and benefits at an OEM dealership.”
- Industry and Company
Include the relevant industry and company details. Specifying the industry helps generate general best practices while including specific company information—such as size, locations, and customer base—can produce more tailored insights and actionable recommendations.
Example: “You are a Sales Manager working in the manufacturing industry, with a focus on material handling equipment.”
- Tone and Personality
Describe the personality and tone you want the responses to convey. The tone should be appropriate for the target audience.
Example: “You are a Customer Support Specialist for an e-commerce retailer. You are friendly and empathetic when interacting with customers, ensuring a positive and supportive experience.”
- Goals and Objectives
Describe the goals and objectives you want the chatbot to achieve. This will guide the development of responses that align with the desired outcome.
Example: “You are a Project Manager with 15 years of experience in the construction industry. Your primary objectives are to ensure projects are completed on time and within budget and to improve project efficiency.”
- Content Format
Specify the format you prefer for the responses, such as a bulleted list for quick reads, step-by-step guides, or full paragraphs for in-depth articles.
Example: “You are a Warehouse Operations Manager. The preferred format of your outputs are step-by-step guides, to ensure clarity and comprehensive procedures.”
- Approach to Unknown
Even with its vast intelligence, the chatbot may not always have the answer. It’s helpful to instruct it to let you know when it doesn’t know something instead of guessing.
Example: “If you don’t know the answer, tell me “I do not know” instead of guessing.”
By clearly defining these key elements, you can create a well-rounded persona that enables the chatbot to deliver more accurate, relevant, and engaging responses.
Tip: Once you’ve created a persona you’re happy with, save a copy so you can easily reuse it later.
Example persona:
Below is an example persona you can try today. Update the wording in [brackets] according to your needs and enter them into your chatbot of choice. Feel free to add as many details as you’d like.
“You are a [job title] with [number of years] of experience in the [industry name] industry and [company details]. You are [desired personality traits and/or tone], and your objectives are to [goals or objectives]. Your preferred format is [content format]. If you don’t know the answer, tell me “I do not know” instead of guessing.”
Completed example:
“You are a Warehouse Manager with 20 years of experience in the logistics and supply chain industry, specializing in small replacement parts. You are authoritative yet approachable, and your primary objective is to optimize the layout of your warehouse, measured by improved packing efficiency of 5% and reduced shrink of 2%. Your preferred format is bulleted lists. Tell me “I do not know” instead of guessing if you don’t know the answer.
Let’s Try It Out
As an experiment, try this prompt without a persona:
“Give me best practices and a detailed how-to guide on warehouse layout optimization for my warehouse.”
Your response will likely be suitable as an introductory guide but will lack the customized, actionable steps that would most benefit your unique warehouse and optimization efforts.
Now, start a new chat, copy in the ‘completed example’ persona above, and add the same prompt immediately after it.
The updated response likely offers a more in-depth and accurate response tailored to your specific industry and operational goals. Moreover, when the chatbot is equipped with a persona, it vastly improves its ability to answer follow-up questions in a way that is relevant to your business needs.
The Bottom Line
AI-powered chatbots can be customized in various ways to meet your unique and business needs. Incorporating a well-defined persona into your prompts is a great way to get more accurate and relevant responses from your AI Chatbot.
As I often advise my clients, education, and experimentation are the keys to kickstarting their AI journey. Try the tech today; your organization will be much further ahead tomorrow.
If you have any questions or need help getting started, please contact me at [email protected]. You can also connect with me on LinkedIn for the latest trends and insights in AI and automation.
About the Author:
Jason Pikoos is a managing partner and leads Connor Group’s technology and innovation, which includes AI-driven solutions. He brings over 20 years of accounting, operational, and technology experience, working with high-growth and technology companies. Jason is a leader in helping companies drive operational excellence through process improvement, technology and automation, data and analytics, and effective governance.
