ServiceNow is on a mission to disrupt the enterprise software industry – to help clients realize exponential gains in productivity and efficiency. ServiceNow is the AI platform for Business Transformation and, with its Now Assist products, it can leverage GenAI to produce improved results and accelerate them.
At Astrica, we are helping clients with this profound shift in both the technology and how organizations think about what we call the ‘surroundings of AI' – the many areas across which AI will necessitate changes. In this blog, we provide one example: Knowledge Management. Knowledge is one of the most powerful capabilities of ServiceNow. Knowledge coupled with Now Assist allows for the ability to use knowledge articles in AI Search and with Now Assist Q&A Genius Results to display concise, actionable answers from knowledge article results in Service Portal, Virtual Agent, Employee Center, and global searches.
Yet there are things to contemplate carefully to ensure the output is what you want and you are managing it well. ServiceNow recommends a set of Architecture and Governance Best Practices when using Knowledge and Now Assist. We highlight some of those below (and please see additional detail provided by ServiceNow here):
Reduce or remove duplicate articles. A Large Language Model (LLM) accuracy depends on the data it's trained on, so it's crucial to keep articles relevant and up-to-date. Maintaining duplicates in various knowledge bases can lead to outdated and inaccurate versions.
💡Astrica Insight: Having duplicate articles can cause a number of issues – if you split article views amongst duplicate versions, your Knowledge managers will be misled on what information is most critical to your end-user base. This could cause not only misuse of time but mis-prioritization of later initiatives.
Currently, Now Assist in Search and Virtual Agent does not summarize content from external sources. To utilize the full benefits of Now Assist within ServiceNow, leverage ServiceNow's Knowledge Management directly.
💡Astrica Insight: These tools work best together – not only is Knowledge the best supported source of content but Now Assist can help your agents generate more of it, continuing a positive feedback loop of expanding resources & higher quality search responses.
Focus on one topic for each article, and cover it fully. The LLM can't click links currently or go to other pages to summarize info. Complete, relevant articles help the LLM summarize better for Now Assist in Virtual Agent and Search.
💡Astrica Insight: Not only is this best for Now Assist but it also can be a UX improvement for your end-users; giving them one place to go for a complete answer improves search results inside and outside of Now Assist.
Write in clear and simple language that is easy to understand. Avoid unnecessary jargon or overly complex language that may confuse the generative AI model or detract from the clarity of the content.
💡Astrica Insight: Remember that you cater to a wide variety of audiences for your knowledge bases – it won’t always be experts reading the articles. Make sure to use language that reaches the broadest number of end-users possible.
Generative AI can parse FAQ articles to some extent, but there are challenges associated with these types of articles. FAQ and Q&A articles may pose challenges with generative AI understanding the context of the article.
💡Astrica Insight: Search in ServiceNow of all kinds is based around more than just the text in the article – one of the best tools in creating good searches is appropriate metadata for the articles. FAQs are great but having separate articles allows you to curate better metadata that helps across the platform.
Generative AI models are typically designed to process and generate text-based content rather than audio or visual content. Therefore, formatting related to audiovisual elements such as timestamps, audio cues, or image descriptions may not be directly processed or preserved by these models.
💡Astrica Insight: This is another good knowledge management procedure whether you’re implementing Now Assist or not – remember that providing appropriate text is a great step in making your knowledge base friendlier to screen-readers and a more gracious experience to all users.
Ensure that content is relevant and up-to-date. Keeping the content up-to-date ensures that the generative AI model generates accurate and relevant responses based on the most current information available. If you're using the Knowledge-Centered Service (KCS) methodology, your articles will be improved via the Solve loop.
💡Astrica Insight: Knowledge-Centered Service (KCS) is turbo-charged with the Now Assist Knowledge generation feature. The highest barrier to effective KCS is often the time it takes to draft quality articles from resolved cases – Now Assist removes that barrier.
Your governance process ensures that the knowledge articles used by Now Assist are relevant, accurate, and reliable. Governance establishes the standards for data integrity and quality, which is essential for generating trustworthy outputs for your end-users and customers. Implement a quality assurance and review process to ensure the accuracy, relevance, and coherence of the knowledge articles. This may involve peer review, expert validation, or automated checks to identify and correct any errors or inconsistencies in the content.
💡Astrica Insight: Knowledge is power – the more effort you provide in its governance and administration, the greater ability for self-service you provide to your users.
Use actionable feedback and feedback functionalities on articles such as rating, flagging, and marking helpful/not helpful for users to provide input on knowledge articles and suggest improvements or corrections. User feedback helps identify areas for enhancement and ensures that the knowledge articles remain valuable and effective for use with generative AI.
💡Astrica Insight: Your agents can only take knowledge so far – without appropriate feedback mechanisms in place and available to your end-users, you’re robbing your organization of the surest way to recognize the value of the content you’ve produced.
In this specific use case of Knowledge, you benefit from new governance approaches, quality oversight, and assurance that the model is being trained to optimize your goals – be they improved customer/employee services, increased reliance on self-service and/or cost efficiencies. Further, as the model evolves it may be important to contemplate more frequent releases into production.
Interested in learning more? Contact Astrica