Understand the Userlane Guides Analytics
First, select the time range that you want to analyze.
How many users started your guides?
Guides started: The number of Guides started could include the same user multiple times if they started more than one guide or started the same guide multiple times.
Unique guides started: The number of users who started one or more guides. Each user will only be counted once towards this number.
Average of unique guides started: The average number of guides started per user. A user starting the same guide can be counted multiple times towards the Guides started number.
Average of unique guides started = Guides started/Unique guides started
This number will indicate the average number of guides one user begins in the selected time frame. Thus, giving you an idea for how often your users engage with your guides.
Where did users start guides from?
Guide source: Where the user found and started the guide.
Assistant menu: the user started a guide via a chapter within the Assistant
Direct link & API command: the user started a guide via a link or an API command
Welcome Slide: the user started a guide directly from the Welcome Slide
Announcements: the user started a guide from an Announcement
Assistant search: the user started a guide after completing a search via the Assistant search
Promotions: the user started a guide from a Promotion
Total: The number of times a guide was started from each source.
%: The percentage of times a guide was started from each source out of the total guide starts.
Evaluate the performance of each individual guide. You will notice that you are able to sort your guides by the different column headers, since the column you choose to sort by will depend on what information you are looking for at the time.
The far left dot will indicate the guide's current status. There will be a green dot if the guide is currently set to public and a grey dot if it is set to private.
Hovering over the icons under the Segments column will give you more information about the segments that have been applied to this guide.
Each guide will have a calculated Health Score. The Health Score indicates how far your users progressed through a guide.
Health Score = [Number of times a guide was completed / (Number of times a guide was started - Number of times a guide was exited on the first step)]
The score ranges from 0 to 100 where a 100 means all users are completing the guide. It is important to note that users can still be receiving value from guides even if they are not completing them. See below for scenarios where this may be the case.
Good to know: There are valid reasons for a low Health Score. Consider if the following scenarios apply to any of your guides:
A guide has a very low Health Score compared to your other guides, and it has an Alert (image above). Users are having trouble completing the guide. Action: Take a look at this article on resolving Alerts. If the guide does not have any alerts, but you know users are being forced to exit the guide rather than doing so intentionally, consider adjusting your Alert parameters. By default, this threshold is set to 5 guide breaks within 8 hours before triggering an Alert. If you want to adjust those values, reach out to your Customer Success Manager.
A guide has a very low Health Score compared to your other guides, but there is no Alert. Users are having trouble completing the guide. If the guide is not breaking, then the users are exiting intentionally. Action: Review the guide to consider why a user might exit. Example: If you included a link within one of the steps in the guide, it is possible that the users are clicking on the link but then never returning to complete the guide. Consider only including links in steps that are towards the end of the guide so that users are not being encouraged to leave the guide before completing important steps.
If a guide has been designed to teach the users how to perform a task rather than assist them while they complete the task, it is possible users will be using this guide as a reference. This means that they may find the answer to a question they had before they reach the end of the guide and will exit early. In this case, the health score may be low, but the user found what they were looking for successfully. Action: Compare the Health Scores of similar guides and consider what the main differences are between these guides. If two guides have very different Health Scores, but you feel they should be performing similarly, it is possible the guide with the lower score should be reviewed for improvement.