You want to improve your digital experience, but how do you know if you've improved it or not? First, you need to know your baseline performance so you have something to compare your future data to. Once you check to make sure everything you need is reporting correctly to our platform, you can validate your information and use our quality foundation dashboard to measure your customer experience.
Our browser monitoring tools are a great way to monitor your app data, and so the best place to start with improving your digital experience is reviewing your browser dashboards. First, you'll need to review your browser apps and pages to make sure that everything you expect to report to New Relic is actually doing so. You can do this by reviewing the Page Views tab in the browser monitoring UI for each app you're ingesting or by running the following NRQL query:
SELECT uniques(pageUrl) from PageView LIMIT MAX
New to NRQL? Check out our introduction to NRQL to learn about how to use our custom query language to help maximize your data!
After you've made sure that your instrumented pages report their data correctly in the previous step, it's time to ensure browser segments are captured correctly so that you can better parse the data that you're using to improve your customer experience. If you're unfamiliar with segments, they're simply the text between two
/s in a URL or the
.s of a domain name.
When you have a lot of URLs with a lot of segments, you can abridge them so that
website.com/product/. While the first version does work, the second version is a more useful way of grouping customer experience data for the product because it allows for more segments in the data.
Not sure whether you need to tune your configuration? Import the Segment allow-list investigation dashboard to help.
Once you've identified all your segments, use the Segment allow-lists in browser to add them, which will enable you to create more easily understood segments.
Now that you've validated your browser URL grouping, you can make your customer experience data easier to understand by breaking it out into different segments. Unlike the previous step, segments here don't refer to sections of URLs (as in the segment allow lists), but instead refer to groups of data.
You can use segments to group your data in different ways. For example:
- Most of the users in the US, Canada, and EMEA experience 2 seconds or better to first input delay, while users in Malaysia and Indonesia experience 4 seconds. Grouping your segments by geographical location would give you this insight.
- Customers buying car insurance typically see 1 second to largest contentful paint. For home insurance, it's 4 seconds.
- In one week, there can be 700,000 page views on mobile browser apps compared to 300,000 on desktop.
Here are popular and useful categories to segment your data:
The next step is to create a dashboard you can use to measure and improve your customer experience against your baseline performance. To do this:
- Go to the Quality foundation quickstart.
- Follow the instructions in the quickstart to install the dashboard.
- Use the provided guide to customize it to fit how you plan to segment the data.
Make sure to align the dashboard to lines of business or customer-facing offerings rather than to teams to maximize the impact of your optimization!
The last step in establishing your current state is to capture your current performance from the dashboard pages. To do this:
- Follow the instructions on our Quality foundation GitHub README, which walk you through the segment allow list and quality foundation dashboard.
- Use the dashboard from the previous step to understand the overall performance for each line of business. If relevant, apply filters to see performance across region or device. If values drop below targets and it matters, add it to the sheet as a candidate for improvement. Examples of tracking-value:
- Not worth tracking: A company that sells insurance in the US only notices poor performance in Malaysia.
- Worth tracking: A company that sells insurance in the US only notices poor performance with respect to mobile users in the US.