AI Search - Operational Limitations and Best Practices

[yoshi0427]

Hello Experts,

I have a question regarding AI Search from an operator's perspective. Are there any operational limitations or considerations we should be aware of?

For example, until a large number of inquiries are accumulated, the search results may not be displayed in the correct order, requiring manual adjustments to the knowledge ranking each time.

I would love to hear about your experiences and any best practices you can share.

Thank you in advance for your insights!

Best regards,

Yoshi

[Dr Atul G- LNG]

Hi @yoshi0427 

Greetings!!

I am implementing AI search with Now Assist for my current client, and while it’s beneficial in some ways, there are operational challenges as well. Even though I’m incorporating Now Assist, the issue is that the AI search returns a lot of knowledge articles (KAs) when searching for a single word, and when multiple words are entered, the results increase even further. This creates confusion for agents when they need to select the best article. The challenge is that agents need to be trained on how to use AI search effectively, understanding that not everything the AI provides is correct. A human mind is still needed to select the most relevant result.

Additionally, to get good results from AI, knowledge articles must be well-written and have properly formatted tags and meta information, which increases the workload for the team. Overall, I can say that we’re still in the early stages of using this tool, and while it takes time to refine, it is certainly helpful.