Google Analytics and Help Center - Part 2: Measuring the effectiveness of search Follow

team professional enterprise plans

You can use Google Analytics to supplement Zendesk Support's built-in search analytics to get additional insights into the effectiveness of search in your Help Center.

The article covers the following topics:

This is the second article in a series. The series outlines how to use Google Analytics to answer questions you may have as a Help Center manager responsible for providing an effective self-service support option to your customers. The series covers the following topics:

Getting started

If you haven't already done so, enable Google Analytics in your Help Center. See Enabling Google Analytics. Once Google Analytics is enabled, turn on site search tracking to start gathering search data.

To enable site search tracking

  1. In the Google Analytics dashboard, navigate to Admin > View > View Settings.
  2. Enable Site Search Tracking.
  3. Enter query as a query parameter.

    When Google Analytics sees query= in a URL in your Help Center, it captures the parameter's value as a search string. For example, the word "guidelines" in the following Help Center results page is captured as the search term:

    https://{subdomain}.zendesk.com/hc/en-us/search?utf8=✓&query=guidelines&commit=Search

Once enabled, search data starts flowing into Google Analytics. It can take a couple hours before the data begins showing up in reports. The rest of this article highlights the metrics to use to measure the effectiveness of search in your Help Center.

Note: If you want to segment your Help Center activity metrics based on user role, see Help Center user segmentation using Google Analytics custom variables.

Do my customers use search?

Use these metrics: Sessions with Search and Percentage of Visits with Site Search

Sessions with Search and Percentage of Visits with Site Search measure the number and percentage of sessions in which the user searched the Help Center at least once during the session. Both metrics help you understand how often customers use search instead of navigating the Help Center using links.

In the Google Analytics dashboard, go to Behavior > Site Search > Overview.

If you want to drive customers to using search as their first course of action, one effective tactic is to display a large, prominent search bar. You can do this with a CSS customization or by selecting the The Wiry Merchant theme, which is designed to emphasize a search-first approach.

Do my customers find answers using search?

Use these metrics:

  • Searches by page
  • Search refinements
  • Search exits by search term

Searches by page

The Total Unique Searches By Page metric is the number of times people searched your Help Center by page. This metric is useful if your site design includes a search box on every article as well as on landing pages. You want to make sure that customers are searching from pages that you'd expect, like the Help Center home page. If they're searching from pages deeper in your Help Center, it might be a sign that they're not landing on the correct content and need to perform another search.

In the Google Analytics dashboard, go to Behavior > Site Search > Pages.

If you see a high number of searches performed from an article, you may want to review the article to make sure it covers what your customers are searching for.

Search refinements

The Search Refinements metric is the percentage of searches that resulted in another search using a different search term. Ideally the metric should be trending down. If a customer has to refine their search multiple times, it means they didn't immediately find a relevant article.

In the Google Analytics dashboard, go to Behavior > Site Search > Overview.

Search exits by search term

The Search Exits and Percentage of Search Exits metrics are the number and percentage of searches a customer made immediately before leaving your Help Center. While the overall metric is displayed on the Site Search > Overview page, a more meaningful metric is the percentage of search exits by search term.

In the Google Analytics dashboard, go to Behavior > Site Search > Search Terms. The metric is a column in the report.

You can measure the popularity of search terms by looking at the total number of unique searches. If popular search terms have a high percentage of search exits, the titles or articles returned for that term may not be optimal. You may want to look into making the article titles more descriptive, or improving the content of the articles. In the example above, "email" was the third most popular search term, but it had a high percentage of search exits.

Are the answers useful?

Use this metric: Time after Search

Time after Search is the average amount of time customers spend on your Help Center after conducting a search. If a customer spends a lot time on the site after performing a search, they probably found something useful. If they didn't spend a lot of time, they probably didn't find anything useful.

In the Google Analytics dashboard, go to Behavior > Site Search > Search Terms. The metric is a column in the report.

Focus on the search terms with a low time-after-search to optimize your search results and knowledge base content. Possible actions include inserting popular search terms as keywords in relevant articles, or creating more content based on the search term.

What answers are my customers not finding with search?

Use this metric: Refined Keyword

Refined Keyword is the search term used on a subsequent search. It's a powerful metric when trying to understand confusion about your product or service. For example, you can see that a user searches for “reporting on articles”, but then refines the search to “how to use google analytics”. It's why this article was written.

In the Google Analytics dashboard, go to Behavior > Site Search > Search Terms, and then add Behavior > Refined Keyword as a secondary dimension.

Have more questions? Submit a request

Comments

  • 1

    Thanks!

    Do I understand correctly that the word "query" is all that's needed? It's not very clear because the font is only slightly different, and the further explanation includes other characters....

  • 0

    *bump*

    Can someone confirm/respond to David's comment?

  • 1

    I can confirm - I entered the word "query" as shown in my previous comment, and now the search data is in fact appearing in the Google Analytics data.

     

Please sign in to leave a comment.

Powered by Zendesk