Knowledge Management

Knowledge Management is the process of creating, curating,  sharing, using and managing an organisation’s knowledge and information.

A Knowledge Base is a self-service library of information about a product, service, department or topic.

Knowledge management is important for businesses as it offers a range of benefits:

• Fosters confidence in employees and increases their satisfaction levels - if employees can easily access documented information, they can answer questions faster, not feel left-out or lacking in information, and will be happier and more productive as a result.
• Improve service delivery - access to accurate knowledge allows service agents to resolve issues faster
• Reduce downtime and disruption - access to quick solutions to common or recurring problems will minimise any downtime
• Faster decision-making - reliable sources of information will allow end-users and agents to get the information they need to make quick decisions
• Knowledge retention - documented knowledge prevents the loss of expertise due to personnel changes
• Service cost reduction - faster incident resolution, self-service incident resolution and shifting service left so that it isn’t escalated to a level 2 or 3 agent help reduce service costs.

Knowledge-Centred Service (KCS) is when service teams not only provide real-time customer, system or employee support, but they also create and maintain documentation as a part of these processes.

For example, if an incident is logged, the service agent will check the knowledge base to see if a fix is documented. If it is, they can easily follow the steps to resolve the incident, taking note to update or change any parts of the article if they’re wrong, outdated or confusing. If there is no documentation, the service agent will follow established processes to resolve the incident while simultaneously documenting the issue and fix in a new knowledge base article.

By capturing information and growing the knowledge base over time, not only will your team be able to respond to incidents more quickly, but you’ll also be building the perfect data set to train Servicely’s native AI so that it can act as both a virtual agent, completing these tasks autonomously, and a copilot for your agents, assisting and providing recommendations to agents in real-time.

Servicely comes equipped with a native Generative AI. Generative AI models use learning models with vast amounts of data to train the AI on how to properly respond to prompts and create content. If an GenAI is trained with generic content, it will give generic answers in response. 

The expertise you can give your AI using more relevant information, the better it will become at responding to prompts. As you use a service management platform, over time the AI will learn from digesting information on problems, incidents, requests and other records. But you can supercharge this learning curve and help your AI become an expert on your business by focussing on your knowledge base.

Creating knowledge articles better enables your AI to:

• Answer specific questions
• Automatically recommend remediations to incidents
• Automate service with self-service via an AI virtual agent

• Knowledge base creation - automatically draft or create knowledge articles by analysing support tickets, customer interactions and documentation. Use prompts to draft articles using AI.
• Provide specific answers to end users - AI can find answers in knowledge articles and documents to answer end-user questions in a conversational format.
• Share relevant knowledge to agents as a copilot - analyse support tickets and surface relevant knowledge articles to aid in their resolution
• Continuously learn and grow

Retrieval-Augmented Generation (RAG) is the process of enhancing the accuracy and reliability of the output of generative AI model by having the AI refer to an authoritative knowledge base outside of it’s training data sources before it generates a response.

RAG helps fill gaps in an LLM by giving the AI access to deeper knowledge on a current or more specific topic.

For example, a LLM without RAG doesn’t know the specific terms of your internal policies, and if a user requested information on their policy-specified entitlements, it wouldn’t be able to answer accurately.

With RAG, Servicely’s AI within your instance has access to your enterprise-specific information, and can refer to and surface this information within the virtual agent, while copiloting tickets, and in a wide range of actions.