Identifying gaps in knowledge through KM Dashboard Reporting

Marcia Josephs
Tera Contributor

On the Knowledge Management Dashboard under the Content Usage tab there is a reported headed User Activity which displays searches and views. Is there anyway to show which search resulted in a KBA view, who carried out the search and their entry point into knowledge (ie Service Portal, Knowledge Base Homepage etc)?

 

There is also a Click Through Rate and a Avg Click Rank but gives percentage change compared to the previous month rather than having the ability to drill into the data behind it. What I'm trying to achieve is to understand where the gaps in knowledge searches are by end users. ie a user searches on the Service Portal or the Knowledge Base Homepage and they receive no results.

 

I want to be able to report on their search phrases and see if there are any common searches and then hopefully be able to identify gaps in knowledge. Once I have this I will attempt to find a Subject Matter Expert who can create content for the missing knowledge.

1 REPLY 1

Leri Andrews
Tera Guru

AI Search Analytics Dashboard can give 'queries with no clicks' and 'queries with no results' which will allow you to see the numbers.  Whether they will be useful is another matter entirely!

 

For example: I see that there were 64 queries of 'family care leave' that resulted in no clicks, I have no way of knowing (with articles on Family Care leave in several countries) which country these people are situated in. Is there a gap? No idea. 

 

Equally I can see 6 queries on 'sabbatical' which resulted in no result. I know there are articles that include this word, but clearly they are restricted by user-criteria to countries where these 6 people are not located and therefore they do not see the articles in their results. However, in their countries perhaps there is an article that would have come back if they had used the word 'personal leave' in the search box. Is there a gap?  No idea. 

 

Hopefully you have a way for your desk agents to flag a knowledge gap? This would be a better route than trying to analyse the search logs.  Or can you do some case analytics?