Best practices: Measuring and optimizing your self-service channels

By guest author Ashley Verrill

As a help desk software reviewer, I am constantly asked for my advice about what buyers can do to ensure they get the most out of their system. My answer is almost always the same -- use every reporting feature possible to foster a perpetual cycle of measurement and improvement.

Self-service channels such as community Knowledge Bases (KBs), are one arena where I’ve noticed specific neglect in regards to measurement. In talking with users, the primary reasons for this are twofold:

• Not knowing what to measure
• Not knowing which actions are needed based on the data you do receive

In this article, I’ll provide answers to both of these questions. By the end of this article you’ll understand how you can use Zendesk to measure and optimize your self-service channels.

This article contains the following sections:

• Measure response rate to identify gaps in service
• Make popular topics easy to find
• Zero activity growth means it's time to create content
• Share how you measure and optimize self-service

Measure response rate to identify gaps in service

Response rate, or the percentage of questions in your community that receive a response, is arguably the most important measure to the success of this support channel. Ideally you want this rate to be 100 percent. To know how you're doing, in Zendesk, you can sort questions in your community to see which ones don't have answers.

The reason why this is so important is because customers will only sign up and use the channel if they perceive their issue as being solved fastest this way. Instead, if they start searching for an answer and find that existing queries haven’t received a response, they will likely not decide to try it themselves. Losing an opportunity to deflect a ticket isn’t the only risk either. You could impact customer satisfaction and loyalty for those people who did post a question and never received a response.

You can use Zendesk ticketing functionality to ensure that every question gets an answer. For example, if a question doesn’t receive a response from the community within a certain amount of time, then an agent can move the question to a ticket so that the team can intervene. The employee responder can either write an answer themselves, or send the question to an active customer community user that’s demonstrated expertise in this arena in the past.

Share how you measure and optimize self-service

Make popular topics easy to find

There are two primary measures that will help you identify your most popular topics; one is page views, and the other is “likes” a KB article or forum answer receives. You can track these measures in Zendesk.

In Zendesk, you can enable Google Analytics to measure your page views. And to measure your "likes," in your help center, you can filter questions in your community by "Votes." 

It’s important that your team identify these popular articles and responses, whether user-created or employee-created, so that you can take efforts to ensure they are found (fast), whether that’s upon arrival to the community homepage, through searching terms in the search bar of the community, or just typing their question into Google.

For self-service users that start from the community homepage, you should make sure your most popular articles are listed on the front page. You can manually choose which articles to show on the home page as Promoted articles.

You will also want to organize these threads into popular topic area buckets. The Zendesk support community, for example, has sections for “Zendesk Apps,” “Product Feedback,” and more.

You’ll notice that some of these sections will remain all of the time. Others will just be used while those topics are popular, when a new feature is released, for example.

To make sure that your articles are found when typed in the search bar, you’ll want to optimize your KB search results. Zendesk Support offers a few tools for doing this, including advanced stemming and implicit phrase queries. To read more about how to use those features see About search engine optimization (SEO) in Help Center.

Making sure that your most popular articles are searchable by Google requires first identifying which terms everyone else used to find that article. Google Analytics in will show you what keywords were used in Google when your page was found.

Zero activity growth means it's time to create content

We’ve already mentioned a couple of metrics you can use to measure the overall level of activity in your self-service channels. One of these was page views, the other was number of “likes.” Others could be number of new users over time and number of new topics created. All of these measures should show show growth over time.

You can take it a step further by calculating your self-service score. The self-service score is defined as the number of users who attempt to use content to solve an issue, divided by the number of users who submit a request for an answer:

Self-service score = Total users of your help center(s) / Total users in tickets

To calculate your score, take the total users of your help center from Google Analytics, and divide it by the total unique requesters in tickets. Compare your score with the Zendesk Benchmark Report discussed in this blog post to see how you stack up against your industry peers.

If your engagement metrics or self-service ratio aren't up to snuff, it more than likely means users aren’t finding the content they’re looking for and you need to create it.

One way to figure out what content is missing is to see what users searched when they arrived on an article, but then still created a ticket. In other words, what were they looking for that they didn’t find in the article you served them. Zendesk provides this data with a measure called “tickets created after search” in Analyzing help center search results. You can also click through and read the notes of the ticket afterwards to figure out what questions should be answered in the new article.

Share how you measure and optimize self-service

This article covers just a few measures you can use to optimize your self-service channels. Check out how customers in our Knowledge Roundtable shared how they measure self-service activity. Add a comment there to let us know what other metrics you use!

If you have a success story about how you improved your metrics, add a comment to the discussion here.