One measurement of how well we’re helping customers answer their questions is tracking our self-service score. It’s a simple ratio of users in our Support Center versus the number of users in support tickets. For example, a self-service score of 3:1 means that for every three people engaging with our documentation, one opens a support ticket.
This score becomes important as LexBlog opens more parts of our platform to more of our customers. We’re giving users more control over settings such as their sidebar widgets and design colors. We’re also rolling out a self-service website model where users can create, design, and launch their own sites.
As people use these new tools, however, they’re going to have questions. And research shows that most customers would rather find answers to their questions before contacting support channels. For example, American Express reports that “48% of customers prefer to speak with a customer service rep when dealing with complex issues, but only 16% prefer the same contact for simple issues.”
Ideally, as we grow, the number of users finding help through our documentation will increase faster than the number of users submitting support requests. Ideally.
Our self-service score over time
The chart above tracks our self-service rate starting in July 2016. That’s when we finalized the transition from the old documentation platform, Reach, to Zendesk’s Guide software. The good news is our self-service rate is increasing over time.
How do we boost the self-service rate?
To be clear: self-service isn’t about discouraging customers from contacting us when they do have trouble. Sometimes when things go wrong you just need to hear a human voice. Our support team is dang good at answering questions and putting out fires.
For me, increasing our self-service score means fewer people get to the point where they can’t figure out something.
And while the increase in LexBlog’s self-service rate is good news, I think it can get better. Here are some ways I want to help our customers through better documentation:
- Continue monitoring search query results reports to see what our users are searching for.
- Deleting content that users don’t read. Unread articles clutter up search results and make it harder for customers to find what they really need.
- Create a smoother onboarding process for new users. This includes guides written for brand-new bloggers.
- Analyzing the path of users through help content. This will tell us where they’re getting stuck.
- Let the robots help. We just implemented Answer Bot, an AI tool that suggests help articles when customers reach out to our support team.
- Getting direct feedback from customers. Future projects may involve getting our customers to provide ideas for our support docs. In fact, if you have any feedback — good, bad, or ugly — on our Support Center, please leave a comment on this post!