Offering self-service options are great, but how can you actually know if your customer service knowledge base is successful?
It’s widely known that customers prefer to solve problems on their own without needing to contact support. What’s not so well known is how to tell if you’re actually good at providing self-service.
The self-service movement has been strong in recent years. According to a study done by Zendesk, 69% of customers try to troubleshoot their own problems before contacting support.
Self-service support not only gives your customers what they want, but it also takes the support load off of agents.
As less-complex tickets are eliminated by customers using your knowledge base, agents become more focused on difficult technical problems.
Simply standing up a knowledge base isn’t enough to meet your customers’ needs. If you build it and then forget about it, you’re going to miss out on opportunities to optimize self-service support.
Even worse, you’re going to create frustration amongst your customers. Having a plan for continuously improving your knowledge base is critical.
But how do you know what needs to be improved? How do you know if your knowledge base is working?
Unfortunately, there isn’t a single way to answer these questions. As you’ll see in this article, you need insight into a number of different metrics to get a good understanding of the performance of your knowledge base.
Comparing those metrics side-by-side will start to reveal valuable insights.
1. Overall Traffic
Understanding how many visits your knowledge base gets is a great place to start. This metric alone isn’t going to give you a ton of clarity, but it’s important to know at a high-level how much traffic your knowledge base is getting so that you can compare this metric with the other metrics below.
For example, if you have thousands of users and hundreds of monthly support tickets but only a couple dozen monthly visits to your knowledge base, you need to know that.
t’s possible your customers just aren’t aware of your knowledge base, which you can change by making it more prominent on your site or sending a message to your customers.
On the other hand, this could also indicate a bug within your knowledge base that needs fixing.
Paying attention to your traffic helps uncover multiple problems and dictates your next steps.
2. Most Read/Unread Articles
As your product evolves, you’ll have an increasing number of articles in your knowledge base. Knowing which of these articles gets the most amount of reads and the least amount of reads can be an indicator of a number of things.
For example, your getting started guide might have a really low number of reads, all while your support team is getting slammed with onboarding questions.
This might be an indicator that your getting started guide needs to be baked into the product or the onboarding process.
3. Support Ticket Volume
Support volume is another metric that isn’t going to give you much insight on its own. But it’s imperative to know ticket volume as you start to look at some of the other key metrics for measuring the success of your customer service knowledge base.
As your knowledge base improves, you should also expect to see the number of support tickets you receive per customer decreasing.
For example, let’s say you saw a huge spike in support volume during a recent feature release. It turns out you shipped a feature but failed to publish supporting documentation for it, so your support team got flooded with tickets.
On a more granular level, you should understand which support inquiries could have been solved with documentation or self-help resources.
If you’re getting a ton of support tickets related to topics covered in your docs, your knowledge base isn’t doing its job.
4. Bounce Rate
Bounce rate is the percentage of viewers who navigate away from your site after viewing just a single page. In other words, these folks walk through the door and turn right back around.
A high bounce rate on your help center means your knowledge base is unsuccessful. It could indicate a poor user experience, poor readability, errors, irrelevant content, or some other issue.
While a high bounce rate should be concerning, you should also remember that if your help center is doing its job some users will find the help they need immediately and then leave the page.
This means a visitor bouncing isn’t always an indicator of an issue with your knowledge base.
If your knowledge base can track visitor behavior, you may be able to report on which users bounce from your knowledge base and then still submit a ticket.
5. Search Metrics
A good knowledge base allows users to search for the content they’re looking for, meaning many knowledge base platforms include search metrics within their reporting.
Google Analytics can also be a great way to capture search metrics within your knowledge base. Search metrics indicate whether or not users are finding what they’re searching for.
For example, if you see a lot of queries for “single sign-on” that aren’t rendering any results, it’s probably time to write some documentation about single sign-on.
Similarly, if you’re seeing a ton of queries for “email customizations” that do get results, this is an indicator that your customers are really interested in email customizations and often need help with it.
It’s probably a good idea to update your existing help articles and fill in any gaps about this topic.
6. Surveys & Customer Feedback
Surveys provide qualitative insights that other metrics and dashboards don’t. Survey your customers annually or bi-annually about their overall support experience, including what they think about your self-help resources.
You’ll be surprised by what you learn. You can use these insights to make a plan for improving your knowledge base.
Many knowledge base platforms also have a simple voting system that enables readers to flag specific content as helpful or not helpful. Use this ongoing feedback to understand the effectiveness of your content and to consistently make improvements.
7. Usability Tests
Just like web designers conduct usability tests to understand how user-friendly a website is, content designers can conduct usability tests to understand how effective their knowledge base is.
Select a handful of active customers and spend 30 minutes with them. Observe how they use your knowledge base. Can they find answers to certain problems?
Do they know how to navigate to your docs from within your application? If they can’t find an answer, can they easily figure out how to contact your support team?
Usability testing can be time-consuming, but it’s a great way to get rich insights about your knowledge base that you won’t find in quantitative metrics.
8. Self-service score
To calculate your self-service score, you’ll need two of the metrics discussed above: traffic and support ticket volume.
Self-service score = Total user sessions of your help center / Total users in tickets
This formula gives you a ratio that helps you understand your help center’s effectiveness.
For instance, if 12,000 users visited your help center last month and 1000 submitted tickets, your ratio would be 12:1.
The best way to measure how successful your knowledge base is requires taking a holistic view.
By tracking and analyzing the metrics shared above, you’ll be blending quantitative and qualitative insights that help you more fully understand how your customers view your help center.
This is how you’ll get useful insights that you can leverage to improve the future success of your knowledge base.
If you’re just getting started with building a knowledge base, check out 5 Practices For Creating a Customer Service Knowledge Base.