User prompts vs. system prompts: What’s the difference?

AI has quickly become the ace up every savvy seller's sleeve. As sellers, we're constantly seeking ways to work smarter, not harder - and AI is our ticket to doing just that. But here's the catch: AI is only as good as the instructions we give it. 

That's where prompts come in.

Prompts are the secret language between us and our AI platforms, the key to unlocking their full potential in our sales processes. Whether you're crafting the perfect cold email or preparing for a high-stakes client call, understanding the nuances of user prompts and system prompts can be the difference between AI-generated mediocrity and sales magic. In this guide, we'll demystify these concepts and show you how to harness their power to revolutionize your sales game. 

Let's dive into the world of AI prompts and explore how they can transform the way we approach sales tasks.

Refresher: What are "prompts" in AI?

A prompt is a set of instructions or queries given to an AI system to guide its output. In essence, prompts are the language we use to communicate with the AI, telling it what we want it to do or what kind of information we're seeking. They can range from simple requests to complex instructions. In the context of sales, prompts can be used to create personalized emails, analyze customer data, generate sales strategies, and much more. Ultimately, they act as a bridge between human intent and machine execution.

As with most things, the quality of your inputs will have a direct effect on the quality of your outputs. In this context, the quality and clarity of your prompts will have a direct impact on the kind of output your AI delivers. Well-crafted prompts can lead to highly relevant and useful outputs, while vague or poorly structured prompts may result in less helpful or off-target responses.

There are two main types of prompts that are particularly relevant for sales professionals using AI: 

• User prompts
• System prompts 

Understanding the distinction between these and how to use them effectively can significantly enhance our ability to leverage AI in our sales processes.

{{cta-1}}

User prompt vs. system prompt: At a glance

Below, we’ve provided a table that highlights the main similarities and differences between user and system prompts.

It’s also worth noting that you may not always need to work with both user and system prompts. For most basic or generally available AI tools like OpenAI’s ChatGPT, Anthropic’s Claude, or Jasper, a user prompt is the only prompt you’d interact with. (For those types of tools, the system prompts would be managed by the platform itself.) 

For more advanced AI solutions like Regie.ai’s Auto-Pilot that can perform automated tasks at scale, you’d be able to adjust both the user and system prompts. This is particularly important when you’re having an AI Agent execute hundreds or thousands of tasks on your team’s behalf; having the ability to adjust both the user and system prompts gives you more control in terms of the outputs your Agents generate. It also gives you the opportunity to make the necessary adjustments needed to refine the output if anything changes in terms of brand messaging, positioning, goals, etc.

What’s a user prompt?

A user prompt is a specific, task-oriented instruction or query that you give to an AI system for a particular interaction. In other words, it’s the "what" you want the AI to do in a given moment. 

User prompts are dynamic and change with each new task or question you pose to the AI. That’s right: we said “task.” You may be most familiar with AI’s ability to generate content, but there are a whole lot of other types of tasks that the technology can handle now. In fact, armed with the right prompts, you can get AI to execute a whole range of sales-related activities, including:

• Creating sales forecasts;
• Comparing competitor feature sets;
• Generating market positioning;
• Conducting SWOT analyses;
• Identifying pipeline bottlenecks;
• Transcribing sales calls;
• Analyzing sales data from a given time period;
• Identifying emerging market trends;
• Analyzing existing customers for churn risk; and,
• Creating detailed ideal customer profiles (ICP) based on your top-performing accounts.

Cool, right? It's like having a brilliant intern who's eager to help but needs clear, specific instructions to deliver exactly what you need. And well-written user prompts help you do exactly that.

How do you write a good user prompt?

Now that we understand what user prompts are and their importance in directing AI output, let's explore how to craft effective ones. While the specific details of your prompt may vary depending on the type of task you’re asking it to execute — whether it's generating cold intro email subject lines or conducting a SWOT analysis — the fundamental principles remain fairly similar across the board.

Let's dive into some key strategies for writing user prompts that will elevate your AI's performance across various sales tasks:

• Be clear and specific about the context: Provide detailed information about the scenario, relevant background, and any constraints. This helps the AI understand the nuances of your request and produce more tailored results.
• Define clear objectives: Clearly articulate what it is that you want to achieve with the AI's output. Whether it's analyzing sales data, creating a strategy, or solving a problem, a well-defined goal guides the AI's focus.
• Break down complex tasks: If your request involves multiple steps or components, break it down into smaller, manageable parts. This helps ensure that each aspect of your request is addressed thoroughly.
• Specify the desired output format: Clearly communicate how you want the information presented, whether it's a bullet-point list, a table, a paragraph format, or something else. This ensures the AI's response is in a usable format for your needs.
• Include relevant data or examples: If you have specific data, examples, or references that are pertinent to your request, include them in your prompt. This provides the AI with concrete information to work with.
• Set any necessary constraints or parameters: If there are limitations, specific criteria, or boundaries the AI should adhere to, make sure to specify these in your prompt. This helps tailor the output to your exact needs.
• Request explanations or rationales: If you want the AI to explain its thinking or provide justifications for its outputs, explicitly ask for this in your prompt. This can be particularly useful for complex analyses or strategic recommendations.
• Encourage creativity within bounds: If appropriate for your task, invite the AI to think creatively or provide alternative perspectives, but within the constraints you've set.
• Specify the level of detail required: Indicate whether you need a high-level overview or an in-depth analysis. This helps ensure the AI's response matches your needs in terms of depth and breadth.
• Ask for specific metrics or KPIs: If your task involves performance analysis or forecasting, specify the key metrics or KPIs you want the AI to focus on.
• Request prioritization or ranking: If applicable, ask the AI to prioritize its findings or recommendations. This can help you focus on the most important or impactful elements of the response.

Remember, the quality of your user prompt directly impacts the quality of the AI's output. The key is to be as clear, specific, and comprehensive as possible while keeping your ultimate goal in mind. Don't hesitate to iterate on your prompts if the initial results aren't quite what you're looking for – refining your prompts is just part of the process.

{{cta-1}}

Examples of great user prompts

Ready to see these principles in action? Below, we’ve provided you with five examples of great user prompts, crafted using the guidelines we’d provided above. Each of these gems illustrates how to craft AI-whispering magic for various sales scenarios. From cold outreach to product launches, these prompts showcase the art of clear instructions, specific contexts, and desired outcomes. Consider this your cheat sheet for becoming a prompt engineering virtuoso in the sales world.

Example #1: Writing a cold intro email 

“Write a cold intro email to a {{JobTitle}} of a mid-sized {{IndustryType}} company. We're introducing our AI-powered {{ProductType}} software. Include the following elements:
- A subject line that mentions AI and a key benefit relevant to the {{IndustryType}} industry
- A brief congratulation on a recent company achievement (use {{RecentAchievement}} placeholder)
- One key statistic about how our software has helped similar companies improve their operations
- A subtle mention of a competitor in their industry who's already using our solution (use {{CompetitorName}} placeholder)
- A soft call-to-action suggesting a 15-minute call to discuss potential fit
- Keep the entire email under 130 words

“Use a professional yet warm tone, and make sure the email is easily scannable. End with a question to encourage a response. Use {{FirstName}}, {{CompanyName}}, and other relevant placeholders throughout the email for personalization.”

Example #2: Conducting a SWOT analysis

“Conduct a SWOT analysis for our company, TechInnovate, a B2B SaaS provider of AI-powered customer service solutions. Focus on:
- Our recent launch of a natural language processing feature
- The increasing demand for automated customer support in e-commerce
- Our challenge in penetrating the enterprise market
- The emergence of new competitors in the AI chatbot space

“Provide 3-4 bullet points for each SWOT category. Conclude with 2-3 actionable recommendations. Limit the analysis to 500 words.”

Example #3: Designing a lead scoring model

“Create a lead scoring model for our company, CloudSecure, which offers cloud security solutions to mid-size businesses. Include:
- 5 demographic/firmographic criteria (e.g., company size, industry)
- 5 behavioral criteria (e.g., website visits, content downloads)
- Point values for each criterion on a scale of 1-10
- Brief explanation for each point value assignment
- Thresholds for lead classification (cold, warm, hot)

“Present the model in a table format with columns for Criteria, Point Value, and Rationale. Conclude with 3 example scenarios applying the scoring model.”

Example #4: Crafting a competitor comparison matrix

“Create a comparison matrix of our product, ProjectPro, against our top 3 competitors in the project management software space: Asana, Trello, and Monday.com. Focus on:
- Key features (list top 5 for each)
- Pricing structure
- Target market
- Unique selling points
- User interface and ease of use
- Integration capabilities

“Present the information in a table format. After the matrix, provide a brief analysis (100 words) of our main competitive advantages and a list of 3 key areas where we can improve.”

Example #5: Creating a new customer segmentation

“Using our B2B SaaS company's customer database, create 4-5 distinct customer segments based on the following criteria:
- Purchasing behavior (e.g., frequency, recency, monetary value)
- Company size (employee count and/or revenue)
- Industry
- Product usage patterns
- Customer lifetime value

“For each identified segment:
- Provide a clear, concise name (e.g., "High-Growth Tech Startups")
- Write a brief description (2-3 sentences) outlining key characteristics
- Suggest 3 tailored sales approaches or strategies
- Identify the primary pain points or challenges this segment likely faces
- Recommend the most suitable products or features for this segment

“Present the segmentation in a structured format, with clear headings for each segment. Include a brief introduction explaining the methodology and potential applications of this segmentation for our sales and marketing teams. Conclude with 2-3 overarching recommendations on how to best leverage this segmentation in our go-to-market strategy.”

{{cta-1}}

What's a system prompt?

A system prompt is a set of overarching instructions that define how the AI should behave across all interactions. If the user prompt is the “what,” the system prompt is the "how" and "why" behind the AI's responses. System prompts are typically set once and remain consistent unless you decide to change the AI's overall behavior or role. They're like the job description and guidelines you give to your AI assistant.

For example, a system prompt for a sales AI might include instructions like: "You are an experienced sales development representative. Always maintain a professional and friendly tone. Focus on the prospect's needs and pain points. Avoid pushy language and prioritize building relationships over immediate sales.

How do you write a good system prompt?

Think of system prompts as the backbone of your AI's behavior and output. While user prompts focus on specific tasks, system prompts set the overall framework for how your AI solution operates across all interactions. 

Let's dive into some key strategies for writing solid system prompts:

• Define the AI's role clearly: Specify the exact role your AI should play, such as a strategic business analyst or a customer segmentation expert. This sets the expertise level and perspective for all interactions.
• Establish the context: Provide a clear understanding of the business environment, industry, and typical scenarios the AI will encounter. This ensures the AI can generate relevant and appropriate responses.
• Set the tone and communication style: Define the overall tone for all outputs, such as professional, analytical, or consultative. This maintains consistency across all interactions and aligns with your brand voice.
• Outline ethical guidelines and compliance standards: Clearly state any legal, ethical, or company-specific guidelines that must be followed. This safeguards against potential issues and maintains professional standards.
• Define the scope of knowledge: Specify the areas of expertise the AI should draw from, such as sales methodologies, market trends, or data analysis techniques.
• Establish output standards: Set universal parameters for the AI's responses, such as the level of detail expected, use of supporting data, or inclusion of actionable recommendations.
• Specify data handling protocols: Provide clear instructions on how to handle and reference sensitive information. This protects privacy and maintains data integrity in all analyses.
• Set problem-solving approaches: Guide the AI on how to approach complex problems, whether it's through data-driven analysis, comparative studies, or strategic frameworks.
• Define collaboration parameters: If the AI will be part of a larger team or process, specify how it should interact with other tools or team members.
• Establish language preferences: Specify preferred terminology, industry jargon to use or avoid, and any company-specific language. This ensures the AI's language aligns with your professional standards.
• Guide on handling uncertainty: Provide instructions on how the AI should handle scenarios where it lacks sufficient information or faces ambiguity.
• Set continuous improvement guidelines: Establish protocols for how the AI should learn from interactions and refine its approach over time.

These guidelines will help you create a system prompt that acts as a comprehensive playbook for your AI. Whether it's analyzing market trends, segmenting customers, or providing strategic recommendations, your AI will consistently deliver high-quality, relevant, and actionable insights that drive your sales goals forward.

Is it really necessary to give your AI a role in your prompt?

Technically, no - it isn't necessary to give your AI a role when you're creating prompts. However, assigning your AI a role can be hugely helpful in effectively communicating with your system and getting the output you're looking for; this is especially true when you’re using AI to create content for an industry like B2B sales, where you're working with a very specific and nuanced lexicon. (And, as you well know, B2B buyers are particularly adept at picking up even the faintest whiff of B.S. or inauthentic language, so there's very little room for error.) By assigning a specific role to the AI, you're essentially providing it with a framework for behavior, knowledge, and decision-making that allows it to more effectively navigate these nuances and create content that truly resonates with your audiences – rather than turning them off.

Let’s dig into this a little more:

1. Contextual understanding: When you give the AI a role, such as "expert sales development representative," it immediately understands the context in which it's operating. This helps it generate more relevant and appropriate responses.
2. Consistent tone and style: A defined role helps maintain a consistent voice across all interactions. Whether it's a "persistent yet tactful sales professional" or a "friendly customer success manager," the AI will adhere to the characteristics of that role.
3. Expertise simulation: Role prompting allows the AI to draw upon a specific set of knowledge and skills associated with that role. This results in more informed and credible outputs.
4. Ethical boundaries: By defining the role, you can also set ethical guidelines and professional standards that the AI should follow, ensuring compliance with industry norms and company policies.
5. Improved personalization: A well-defined role helps the AI understand how to interact with different types of prospects or customers, leading to more personalized and effective communication.
6. Goal alignment: When the AI understands its role, it can better align its outputs with the overall objectives of that role, whether it's generating leads, closing sales, or providing customer support.
7. Enhanced creativity within constraints: Paradoxically, by giving the AI a specific role, you're enabling it to be more creative within the boundaries of that role, often leading to more nuanced and situation-appropriate responses.

By leveraging role prompting, you're essentially giving your AI a persona to embody, complete with its own set of skills, knowledge, and behavioral guidelines. This results in more coherent, purposeful, and effective AI-generated content for your sales communications.

{{cta-1}}

Examples of great system prompts

Now that we've explored the key elements of crafting effective system prompts, let's dive into some concrete examples. The following prompts showcase how to apply these principles across various sales scenarios; specifically, they demonstrate how to set clear guidelines, establish the right tone, and provide specific instructions for AI-generated content. These examples can serve as templates that you can adapt and refine for your unique sales needs.

Example #1: Writing a cold intro email

“You are an expert sales development representative specializing in crafting compelling B2B cold outreach emails. Adhere to these guidelines in all your email compositions:
- Maintain a professional, confident tone without being pushy or overly familiar
- Focus on the prospect's potential pain points and how your solution addresses them
- Keep emails concise, aiming for 150 words or less
- Use simple, jargon-free language unless the industry specifically requires technical terms
- Personalize each email based on the prospect's role, industry, and company
- Include a clear, low-pressure call-to-action
- Avoid making unsubstantiated claims or promises
- Structure emails for easy scanning, using short paragraphs and bullet points where appropriate
- Craft subject lines that are intriguing but not clickbait-y, ideally under 50 characters
- Always provide value in the email, whether it's an insight, resource, or relevant case study
- Incorporate social proof when possible, such as relevant client success stories or industry recognition
- Ensure all content complies with anti-spam regulations and email best practices
- Use placeholders (e.g., {{FirstName}}, {{CompanyName}}) for recipient-specific information

“Your cold emails should be engaging, relevant, and focused on starting a conversation rather than making an immediate sale.”

Example #2: Conducting a SWOT analysis

“You are an experienced business strategy consultant specializing in the B2B SaaS industry. Your role is to provide insightful, actionable SWOT analyses that help companies improve their market position. You have a deep understanding of current market trends, competitive landscapes, and growth strategies in the tech sector.

“When conducting SWOT analyses:
- Always maintain a neutral, objective tone
- Base your insights on current market data and trends
- Prioritize actionable items over general observations
- Consider both short-term and long-term implications
- Balance internal factors (strengths and weaknesses) with external factors (opportunities and threats)
- Use clear, concise language appropriate for executive-level discussions
- Avoid jargon unless it's industry-standard terminology

“Your analyses should be structured, easy to read, and directly applicable to strategic decision-making. When suggesting actions or strategies, always consider their feasibility and potential impact on the company's resources and market position.”

Example #3: Designing a lead scoring model

“You are an experienced sales operations analyst with expertise in lead qualification for B2B tech companies. When designing lead scoring models, adhere to these principles:
- Prioritize criteria that have proven correlation with higher conversion rates
- Use clear, non-technical language accessible to both sales and marketing teams
- Ensure models are flexible and can be easily adjusted based on new data
- Consider both explicit (demographic, firmographic) and implicit (behavioral) factors
- Include negative scoring factors to identify less promising leads
- Base point values on data-driven insights and industry benchmarks
- Design models that integrate with common CRM and marketing automation platforms
- Include time-decay factors for behavioral scores to prioritize recent activities
- Recommend periodic review and recalibration processes for the model
- Suggest methods for testing and validating the model's effectiveness

“Your lead scoring models should be comprehensive, practical to implement, and directly tied to improving sales efficiency and conversion rates.”

Example #4: Crafting a competitor comparison matrix

“You are a market research analyst specializing in competitive analysis for the software industry. When creating comparison matrices, follow these guidelines:
- Maintain strict objectivity in all assessments
- Use consistent criteria across all compared products
- Present information in a clear, easily digestible format
- Use industry-standard terminology where appropriate
- Base all comparisons on verifiable, up-to-date information
- Highlight key differentiators for each product
- Include pricing information and any available trial or free tier options
- Consider user reviews and ratings from reputable sources
- Evaluate integration capabilities and ecosystem compatibility
- Assess customer support options and service level agreements
- Include information on market share and customer base size if available
- Consider future roadmap and recent feature releases

“Your matrices should facilitate informed decision-making by providing a comprehensive, unbiased comparison of key product features and company characteristics.”

Example #5: Creating a new customer segmentation

“You are an advanced customer segmentation analyst specializing in B2B markets. Your role is to create meaningful, actionable customer segments based on various data points. Adhere to these principles in all your segmentation analyses:
- Use data-driven approaches to identify distinct, meaningful segments
- Ensure segments are mutually exclusive and collectively exhaustive
- Balance quantitative metrics with qualitative insights for rich segment profiles
- Consider both current value and potential future value in your segmentation
- Use clear, concise language to describe each segment
- Provide actionable insights that can directly inform sales and marketing strategies
- Ensure segmentation models are scalable and can be updated with new data
- Consider the customer lifecycle in your segmentation approach
- Identify key differentiators between segments
- Suggest tailored approaches that align with each segment's characteristics and needs
- Use industry-standard terminology and avoid jargon unless necessary
- Consider both demographic and behavioral factors in your segmentation

“Your segmentation should provide a clear, comprehensive view of the customer base that can guide strategic decision-making and personalized engagement strategies.”

Final thoughts

As we continue to integrate AI into our sales processes, mastering the art of prompts will become increasingly crucial. By understanding the nuances of prompt engineering, we’ll be able to unlock the full potential of AI, creating a powerful synergy between human expertise and machine efficiency that can dramatically enhance our sales outcomes.

The key lies in continuous learning and adaptation. As you experiment with different prompts and refine your approach, you'll discover new ways to leverage AI for everything from data analysis to content creation. By harnessing AI’s capabilities effectively, we can focus more on what truly matters in sales: building meaningful relationships and delivering value to our clients. So, let's embrace this powerful tool with enthusiasm and creativity, pushing the boundaries of what's possible in AI-assisted sales and staying ahead in our competitive landscape.