Guide to the Self-Service Dashboard

  • 5 February 2021
  • 11 replies
  • 626 views

Badge

Currently the self-service dashboard only supports tickets where Zendesk is the source.

Let us know in the replies if you use another Helpdesk and would also like to report on your self-service performance with this dashboard.

The self-service dashboard shows the proportion of your customers that are able to help themselves with help content, without having to talk to your support team via a ticket.

With this dashboard, you can see whether you’re enabling greater self-sufficiency amongst your customer base, or whether customers are still spending a lot of time dealing with the back-and-forth of support.

Value metric: Self-service score

The value metric for the self-service dashboard is self-service score. We believe that this metric is a leading indicator for NPS (users can solve problems quickly, without having to wait for a support agent). Plus, you’ll save on the time that your support team needs to spend on one-to-many issues, giving them the space to help customers with 1-1 problems that need human intervention and to support customers on a more tactical level.

The calculation is: Total unique visitors who viewed help content / Total unique users who created tickets. A good self-service score depends on your industry (for instance, B2B companies tend to have a higher self-service score than B2C companies) - see this article by Zendesk to go deeper on this.

Metric reference: Self-service dashboard

This table explains in detail how each metric shown on the self-service dashboard behaves, and any limitations or behaviours to watch out for.

 

Metric Description Extra Details
Self-service score The ratio of unique visitors who viewed help content to unique users who created a ticket in the selected time period.

Calculation: count of unique users who viewed self-service content / count of unique users who submitted support tickets.

Ticket data is imported via your external helpdesk software.

Unique visitors who viewed help content The number of unique visitors who viewed an article, question with a best answer, or a conversation on the community in the selected time period.

Page views on topics that have been trashed/marked as spam not considered when calculating unique visitors.

We make an assumption that conversations/articles are more likely to help users than a question without a best answer marked. However, when a question has a best answer marked, then all views on that question are counted, even those before the answer was marked.

Both logged-in and anonymous visitors are included. Visits from different devices are counted as separate, unique visitors. 

Page views from bots are automatically excluded when calculating unique visitors.

Unique visitors who viewed help content % change The percentage difference between the number of unique visitors who viewed help content in the current/previous time period. -
Unique users who created tickets The number of unique users who created a support ticket in the selected time period. Ticket data is imported via your external helpdesk software.
Unique users who created tickets % change The percentage difference between the number of unique visitors who created tickets in the current/previous time period. -
Self-service and helpdesk user trends Unique visitors who viewed help content and unique users who created tickets over time, shown on a daily interval.  

 

Filters

Use the following filters to adjust how the ‘unique visitors who viewed help content’ metric is calculated (and therefore your self-service score).

  • Published in: choose which categories and/or groups should be included when calculating unique visitors.
  • Content type: choose which content types should be included when calculating unique visitors  (available options: article, answered question, conversation).
  • User role: choose which primary or custom user roles should be included when calculating unique visitors.

11 replies

Badge

Both logged-in and anonymous visitors are included. Visits from different devices are counted as separate, unique visitors. 

 

@daniel.boon thank you for this! Is there a way to report on unique registered users who viewed help content within a certain period? i.e. I’d like to see how many of the total unique users who viewed help content were registered users, and subsequently would like to see who those registered users are. Simply being able to report on “content views” of all users as an activity within a certain period could help here too. 

 

The use case here is that we’d like to compare the individuals who are making up our self-service score in community + Zendesk to see if they are the same or different users. We are seeing our self-service score improve, but it is only a result of our community activity going up, not our ticket count going down. Ideally we’d like to see a downward trend of support tickets over time, and I believe the best way to do that would be to promote the community to those who most frequently submit support tickets but might not be active in the community.

 

Thanks!

 

Badge

That's a really interesting question / use case @adam.ballhaussen 

How does the InSided/Zendesk know who has or hasn't raised a ticket? Is it all tracked via IPs? 

Badge

@Gabolino we capture email with all of our support tickets so we are able to identify every user who submits a ticket. I assume that’s what inSided is using via the Zendesk integration to know how many unique users have submitted a ticket.

 

I am hoping that the “viewed help content” information lives somewhere on registered users so that we could identify which registered users viewed content, thus allowing us to compare to our Zendesk users who are submitting tickets. 

Badge

We do to! (capture emails, I mean). But, for all those who are unlogged, neither InSided nor Zendesk knows their emails, so Im assuming they are using IPs as a way to know whether or not they have landed a ticket in support. 

Hope @daniel.boon can clarify.

Badge

@Gabolino yes I believe you’re correct. They’re probably using IPs to calculate total number of visitors. I believe they’re only doing so for inSided though. I don’t think they’re calculating any “visitor” users for Zendesk since this self-service score is only measuring views of help content within the community. I could be wrong though

Badge

Makes sense @adam.ballhaussen I just enabled this a few days ago and my understanding is that it's looking a unique community visits / total tickets created in Zendesk. I can see why you want to isolate registered users here. 

If the above is correct my problem with this score is that there doesn't seem to be a real correlation between those who visited the community and those who created a ticket  - is it just looking at total numbers independently on both platforms? 

The Zendesk Help Center (Guide) gives you visibility of tickets created after visiting an Help Center article. So you can see number of tickets created after visiting the Help Center as well as which articles deflect more (or less) tickets. Furthermore, support folks know when  tickets come from the help center and from which piece of content. 

 

 

Badge

Is there a way to report on unique registered users who viewed help content within a certain period? 

@adam.ballhaussen - this is not currently possible, but I like your thinking, to identify which users are ‘heavyweight’ ticket submitters who could be utilising the community more.

We use cookies to track unique visitors on inSided (since not every customer viewing help content is logged-in), and then we do indeed compare these unique visitors directly to unique email addresses from Zendesk. So yes @Gabolino this tells you about the distribution of activity across channels, rather than performing any linking/showing any correlation between user behaviour across the two platforms.

Badge

I understand. I guess what we all want to know is: if you asked a question in the community and/or logged in, did you create a tickets within let's say 7 days after your visit? That would be useful and something to report on internally. Currently it's more a comparison of views vs tickets which to be totally honest with you is not evidence of ticket deflection. 

Badge

Yes that’s right 🙂 The self-service dashboard in its current state does not aim to show ticket deflection, rather it aims to show you the distribution of engagement across two channels: your community + your ticketing software.

Showing ‘true’ ticket deflection via inSided would be possible if we could track across users both inSided + Zendesk the ticket creation flow → email address helps somewhat towards this, but it won’t account for users who visit inSided anonymously and then submit a ticket on Zendesk. Unless we had the support ticket creation flow embedded directly within a page on inSided :)

Userlevel 5
Badge +4

Add me to the list of people interested in what @adam.ballhaussen and @Gabolino discuss in this thread. If there’s an idea for it, I’ll vote for it and hope it’s considered for future development. It’d be a killer metric for calculating/demonstrating ROI. (Current version is good too, but the unique-to-unique match would be an amazing proof point beyond the assumption within the self-service ratio score.

Otherwise, this was a very helpful article @daniel.boon . One of the best I’ve read on here. Well done. 

Badge

@daniel.boon I hope you’ve been well! I saw the recent tracking improvements to the Audience Dashboard (kudos @Alex Campos 👏)

 

 

The ability to track Registered unique visitors vs. Guest unique visitors on the Audience dashboard got me thinking… could this same tracking functionality be applied to the Self-service dashboard to help us achieve the use case we outlined in this thread?

 

My thoughts

The Self-service Dashboard currently displays the following data:

  • Self-service score
  • Unique visitors who viewed help content
  • Unique users who created tickets
  • Self-service score monthly trend (for selected time period)
  • Self-service and helpdesk user daily trends (for selected time period)

It would be great if the Self-service dashboard could also show trends for registered users vs. non-registered users, similar to what’s now available in the Audience dashboard:

  • Self-service score for registered users
  • Unique registered visitors who viewed help content
  • Unique registered users who created tickets
  • Self-service score monthly trend for registered users (for selected time period
  • Self-service and helpdesk user daily trends for registered users (for selected time period)
  • Self-service score for non-registered users
  • Unique non-registered visitors who viewed help content
  • Unique non-registered users who created tickets
  • Self-service score monthly trend for non-registered users (for selected time period
  • Self-service and helpdesk user daily trends for non-registered users (for selected time period)

This would help us understand behavior of customers who have registered for the community vs. customers who haven’t to identify trends in each of these audiences and determine where we should be focusing our efforts.

 

Current self-service dashboard
Current audience dashboard

I know it’s a bit duplicative, but I created an idea where I copied my thoughts from above. Here’s the link:

 

 

@Gabolino @DannyPancratz what do y’all think?

 

CC: @erin.brisson 

Reply