"Hello, this wiki will provide insights into Axero Copilot, along with best practices for effectively implementing me among your users."

What is Axero Copilot?

Axero Copilot is an advanced AI-powered chatbot built on the foundation of OpenAI's GPT technology.  Axero Copilot is uniquely designed to operate within your organization's intranet, providing a private and secure environment for accessing and interacting with internal content. 

Unlike traditional search engines or knowledge management systems, Axero Copilot goes beyond merely retrieving relevant documents; it comprehensively reads and analyzes content to provide concise and accurate responses to user queries. By utilizing natural language processing, Axero Copilot enables users to engage in conversational interactions, significantly enhancing knowledge discovery and retrieval within the organization.  

Axero Copilot Features & Capabilities

Search Results: Axero Copilot is designed to search for keywords and read unfiltered documentation results to provide answers efficiently. 

Unfiltered documentation search results are derived from the following content types:

• Articles
• Wikis
• Blogs

Excluded content types are:

• Events
• Tasks
• Cases
• Recognition

While Axero Copilot's primary function is to search documentation content effectively, it may also read and understand:

• A subset of user profile data
• Some text page builder pages

Access to Intranet Content: Axero Copilot has exclusive access to your organization's intranet content, ensuring that all interactions occur within a secure environment. It can search through a wide array of documents, including articles, presentations, and files, providing users with instant access to valuable information.

Conversational Interface: Unlike traditional search interfaces, Axero Copilot offers a conversational experience similar to interacting with a colleague or teammate. Users can ask questions naturally, without the need to navigate complex search queries, facilitating smoother knowledge retrieval processes.  

Personalization and Permissions: Axero Copilot operates on a personalized and permissioned basis, ensuring that users only access content relevant to their roles and permissions within the organization. This personalized approach enhances data security and confidentiality.

Content Analysis and Relevance Scoring: Axero Copilot uses logic to analyze content and assign relevance scores to retrieved information. This enables it to prioritize and present the most pertinent details to users, streamlining the information retrieval process.  

Customization Options: Each Axero Copilot installation can be customized to align with the specific requirements and branding preferences of your organization. From the interface design to the naming conventions and logo, customization options ensure seamless integration with existing systems and workflows. 

How Axero Copilot Works

When a user asks a question, Axero Copilot locates the relevant information and responds with a concise summary in text format. Below the line, Axero Copilot primarily targets content providing insights into the response to user queries. It includes relevance scores and references to similar topics, all tailored based on user permissions to ensure access to authorized content only.

• Search and Retrieval: When a user poses a question, Axero Copilot conducts a comprehensive search across all intranet content to find relevant information.
• Content Analysis: Axero Copilot reads and analyzes the content of retrieved documents, identifying key details and extracting relevant information.
• Conversational Interaction: Users engage with Axero Copilot in a conversational manner, asking questions and receiving concise responses based on the analyzed content.
• Relevance Scoring: Axero Copilot assigns relevance scores to retrieved information, prioritizing the most important details for user consumption. 
• Feedback Option: After receiving an answer, users can provide feedback. If the response wasn't helpful, they can specify the reason and provide additional details before submitting their feedback.

Are Customizations Possible?

We know every organization is unique, and Axero Copilot can be customized to fit your needs. Here are some things that can be changed:

• Appearance: We can change how Axero Copilot looks to match the organization's style.
• Name and Avatar: We can give Axero Copilot a unique name and picture to make it match your brand and improve the UX.
• UI Color: If you have specific colors to use from marketing, we can adjust Axero Copilot's interface to match.

Axero Copilot Analytics

Analytics provides administrators with invaluable insights into Axero Copilot user interactions, enabling them to optimize performance and enhance user satisfaction. With sorting, filtering, customizable timeframes, and detailed session information, administrators can review activity, analyze feedback, and understand user behavior to optimize the Copilot's performance and enhance user satisfaction.

Learn more about Axero Copilot Analytics here

Axero Copilot Best Practices

Here are some best practices to help guide your organization with the roll-out of Axero Copilot among your users. These recommendations will optimize user adoption and enhance the overall effectiveness of the Axero Copilot Implementation process.

Content Size Optimization

Axero Copilot is a master at reading a lot of content quickly, but it too has limits in how much information it can read before synthesizing a response. If you find your Axero Copilot is having a hard time digesting large files and documents; connect with your implementation specialist to identify the best strategy to help your users get the responses they need. It may be we can adjust prompts to focus the AI's attention; upgrade GPT models for deeper document reading, or employ other strategies to make content more digestible; such as splitting up PDFs with hundreds of pages. 

Formulating Effective Queries

We encourage users to ask specific questions in natural language. Instead of vague inquiries, we encourage users to be precise and provide context where necessary. Include relevant keywords or phrases to help Axero Copilot better understand your query and provide accurate responses.

• Here is an example of formulating an effective query for Axero Copilot:
  • Ineffective Query: "Budget report"
    • The ineffective query, "budget report" is quite general and could encompass a wide range of references or information.
  • Effective Query: "Can you provide the budget report for Q1 2024 for the Finance department?"
    • The effective query specifies the time period (Q3 2023) and the department (Finance), providing clear parameters for Axero Copilot to search within.

Narrowing/Focusing Responses

When users must follow up on initial responses to gain more focused or detailed information. They can do this by asking clarifying questions or providing additional context to steer the conversation in the desired direction.

• Here's an example of narrowing or focusing a response after receiving an initial answer from Axero Copilot:
  • Initial Query
    • User: "How do I set up a virtual meeting using our company's video conferencing software?"
  • Initial Response
    • Axero Copilot: "To set up a virtual meeting using our company's video conferencing software, follow these steps: 

1) Log in to the software using your company credentials.

2) Navigate to the 'Schedule Meeting' option.

3) Enter the meeting details, including the date, time, and participants.

4) Choose the desired settings, such as audio and video preferences.

5) Click 'Schedule' to confirm the meeting."

• • Narrowing/Focusing Response: 
    • User: "Thanks for the steps. Could you also clarify how to invite external participants to the meeting?"
      • By asking a follow-up question to clarify a specific aspect of the initial response, the user can receive targeted information that directly addresses their needs. In this example, the user is seeking additional guidance on inviting external participants to the virtual meeting, allowing Axero Copilot to provide further instructions or tips tailored to this particular requirement. 

Understanding Responses

When discussing Axero Copilot with your users, emphasize the importance of paying attention to the relevance score assigned to each response, as higher scores signify greater alignment with the query. 

• Here's an example of the relevance score (r:90) from Axero Copilot:

Providing Feedback and Types of Useful Feedback

We encourage users to regularly provide feedback following interactions with Axero Copilot, as it can offer insights for enhancing the performance and accuracy progressively. Instruct your users to specify the reason for their feedback, whether it pertains to identifying inaccuracies, suggesting improvements, or expressing satisfaction with the provided responses. 

• Here's an example of Axero Copilot feedback interface: 
  • Teach users to distinguish between feedback on the content of the response (accuracy, relevance) and feedback on the functionality or user experience of Axero Copilot (ease of use, speed). 

• Here's an example of good and bad feedback:
  • Helpful Feedback: "The response provided by Axero Copilot was accurate and helped me quickly find the information I needed. It precisely addressed my query and saved me a lot of time."
    • Why it's helpful: This feedback highlights the accuracy of the response, specifies how it helped, and acknowledges the time-saving aspect of using Axero Copilot.
  • Unhelpful Feedback: "Axero Copilot's response was inaccurate. It didn't address the main point I was asking about. This tool is useless."
    • Why it's unhelpful: This feedback only states that the response was inaccurate without specifying what was the issue with the response or providing any details about the query. It also uses a strong adjective ("useless") without offering suggestions for improvement.

What your Axero Copilot Can't Do

Understanding what Axero Copilot can't do is essential for your organization to manage expectations and maximize its utility. While Axero Copilot is a capable assistant, it's important to acknowledge its limitations—such as making your coffee! Axero Copilot may not always provide  contextually refined responses. Some scenarios where it can't do its magic include:

Lack of Documented Procedures

• User Query: "How can I troubleshoot an issue with my Lenovo ThinkPad P14s workstation?"
• If the troubleshooting procedures for the technical issue are not adequately documented within your Axero site, Axero Copilot will not be able to provide detailed or comprehensive guidance. Incomplete or outdated documentation in your Axero site will limit Axero Copilot's ability to retrieve relevant troubleshooting steps, leading to unsatisfactory answers for users seeking assistance with technical issues. 

Insufficient Coverage of Specialized Topics

• User Query: "What are the best practices for communicating with the IT department?" 
• If your Axero site lacks comprehensive coverage of specialized topics like communicating with IT, Axero Copilot will struggle to provide satisfactory answers to queries about best practices. Limited or outdated content related to specialized topics within your Axero site can hinder Axero Copilot's ability to retrieve relevant information, resulting in less helpful responses for users seeking guidance on specific practices.

Highly Contextual or Personalized Queries

• User Query: "What tasks should I prioritize for my role as a marketing manager in the upcoming quarter?" 
• Axero Copilot will struggle to provide personalized task prioritization recommendations tailored to your users specific role and responsibilities within your organization. Such queries require a deeper understanding of your user's individual context and priorities, which is not possible for Axero Copilot to discern only from your intranet site content. 

Ambiguous or Unclear Queries

• User Query: "Tell me about the latest content."
• The query "latest content" is ambiguous and lacks specificity, making it challenging for Axero Copilot to determine the intended topic of interest. Without additional context or clarification, Axero Copilot will provide generic or broad responses that may not fully address the query. 

Sensitive or Confidential information Queries

• User Query: "What is my salary?" 
• Axero Copilot will not have access to or be authorized to retrieve sensitive or confidential information such as salary details. Queries that involve such information may be outside the scope of Axero Copilot's capabilities and access permissions, as retrieving such information could pose privacy and security risks. 

In these situations, you may need to seek assistance from human experts or consult external resources to obtain satisfactory answers. Understanding these boundaries help our users to use Axero Copilot effectively while recognizing its strengths and knowing when to seek additional support, even if it's just for a caffeine boost!