Introduction to Metrics

Metrics are a way for you to track key interactions and KPIs on your site or mobile application. Metrics can be plotted in Dashboards and sliced and diced in the Metrics section of the UI.

What is a Metric?

Metrics are an Event (click, page visit, error, or Custom Event), aggregation type and a format (%, $, etc). The aggregations available include count of the event itself, the number of users who performed the event, the number of sessions in which the event was performed, or a calculation of Custom Event properties. 

Metrics give you the power and flexibility to track the specific digital experience interactions that are meaningful to your business. To start building metrics, you can either:

• Select it from the Library on the left side of your screen

OR

• Click Track a Metric from your Home page to instantly be dropped into the Metric builder.

Types of Metrics and how to use them

For all events

• Count of Unique Users - the unique number of users to perform the event (learn more about how FullStory determines a user)
• Count of Events - the number of times the event happened
• Count of Unique Sessions - the number of sessions in which the event happened
• Count of Unique Property - the number of unique properties for a specific event or user variable. (for example, a count of the unique accounts - property: AccountName - that performed a specific event.)

Note: unique properties are limited to properties sent via API for a userVars or custom events.

Example:

• Session 1: User A from Account 1 clicks Add to Cart 3 times
• Session 2: User B from Account 1 clicks Add to Cart once
• Session 3: User C from Account 2 visits the homepage and bounces
• Session 4: User A from Account 2 clicks Add to Cart twice

If your event looks like this:

Then your calculations will return:

For numeric event properties

If you are sending Custom Events with numeric event properties to FullStory, then you will have many new ways to analyze them with Dashboards. The aggregation types highlighted below are specifically for these types of event properties, and not the events themselves (that is where you use Users/Events/Sessions as described above).

Property aggregated by

Below we'll look at what each property in the above dropdown represents and a helpful example.

• Average - The sum of a set of numbers divided by the count of those numbers for a particular metric.
  • Example - When calculating average session duration, the value obtained represents the typical length of time users spend on your site or app during a single session.
• Median - Represents the middle value of a particular metric.
  • Example - When calculating median session duration, the metric would show the middle value of all captured session durations.
• Percentile - Calculates the value below which a percentage of your data falls. Using percentiles is standard practice for effectively measuring site performance, but you can use percentiles to help avoid data skew for any uneven distributions. Choose between 75, 90, 95 and 99th percentile.
  • Example - When looking at a percentile metric for page load times, the 95th percentile value indicates the duration that is faster than or equal to 95% of captured page load times.
• Sum - Total aggregated value of a particular metric or attribute measured over a specific period.
  • Example - When calculating Page Views, the sum would be the total number of views captured over a specific timeframe.
• Min -  The minimum value observed for a specific metric.
  • Example - When calculating Time on Page, the minimum value would indicate the shortest duration of time that a user spent on a particular page during their session.
• Max - The maximum value observed for a specific metric.
  • Example - When calculating Time on Page, the maximum value would indicate the longest duration of time that a user spent on a particular page during their session.

When you select one of these aggregation types, you will be required to choose the 'Property' that you want to calculate:

In this case above, we have a Custom Event called 'Order Completed' that has three numeric properties that describe the event (revenue associated with the order, the shipping, and the tax). By selecting the 'Average' and 'Revenue' I can track my Average Order Value in FullStory.

Not sending any Custom Events to FullStory?

No problem, you can still use these aggregation types on the Event Properties FullStory is collecting automatically. A popular one for our customers is to track the average page load time using data captured by FullStory.

All time related data that FullStory captures are in milliseconds (ms) so we automatically select that format for you. We will also automatically autoscale the display on the visualization to our best time value guess, but clicking off Autoscale will keep the visualization in milliseconds.

The configuration above would create a Metric reflecting the average First Contentful Paint time experienced by users who visited the`/cart` page of our example site. We can visualize this with a Trend card to track page speed over time:

You may find that some metrics are more complex than just a simple count. In that case, use the Add an operator dropdown to perform basic mathematical operations when you need to show a relationship between different events, like ratios and percentages.

Digging into a Metric

FullStory lets you explore the Metric by plotting different Segments of users, Dimensions like Browser Type or Element/CSS Selector, Time Ranges and pivoting between various trend charts.

Adding or Comparing Segments

Keep in mind when applying segments, if new users are included in these, it will not be able to show past data as there won't be anything to report.

Group by Dimensions

If you Group by Element or CSS Selector, FullStory aggregates all Named Elements into single elements, removing all potential duplicates. You also have the ability to Hide CSS Selectors from your results via checkbox.

When you adjust these toggles on the data visualization, you are not altering a saved metric i.e. if you come back later, your Dimension and Time range will be reset. If you find something interesting while exploring a Metric, we recommend creating a Dashboard Card with those settings. 

Time Range and Comparing to past

Additionally, you have the option to visualize the data that is important to you in a Period-Over-Period analysis by using the Compare to past button. For example, maybe you're interested in weekly site visits based on a specific campaign you're running or how a promotional holiday sale performed this year versus the year prior. With Compare to past, FullStory offers the tool for you to see this data up close and learn what worked before and/or what is working now.

Start by simply clicking the Compare to past button and selecting your date range. The date picker will show two colors representing the date ranges you are comparing - blue for current and red for previous. Once you've made your selections, click apply to see the results displayed in a trend chart.

You will see two superimposed lines overlapping that represent each date range. Hover over the dots to see further detail.

Trend Charts - Standard, Rolling Average and Rolling Window

Switch between Standard, Rolling Average or Rolling Window to visualize data points. Use these to calculate common trends such as Daily Active Users, Weekly Active Users or Monthly Active Users and add them to a dashboard. The Rolling Average & Window views provide a more stable time series analysis by smoothing out natural fluctuations in your data, providing a clearer picture of underlying patterns to help you make informed decisions.

• Standard: The standard chart displays static data over a period of time.
• Rolling Average: This chart averages the data over a distinct period of time.
• Rolling Window: This chart aggregates the data over a distinct period of time to view data sets within a specific window of time.

Combining Quantitative Metrics with Qualitative Sessions

FullStory will provide the sessions applicable to Metrics right below the data visualization. If you add an operator (e.g. ÷ × + −) to your metric, you can use the Event 1 and Event 2 toggles to focus on which experience you want to watch in a session replay. If you click on a session, it will take you right to the point in playback that the Event in your Metric occurs. This is important to (1) validate that your metric is capturing the interaction you care about and (2) understand why something might be happening.

Saving a Metric to a Dashboard or Home

Once a Metric has been saved in the 'Metrics' section of the UI, all users can add it directly to Cards in Dashboards or Home. This can help ensure all users are using consistent Metrics when reporting data.

You can also add a metric to a dashboard by using the Add from Library function.

Need to adjust a Metric slightly? No problem. Click edit on any dashboard card and adjust the Metric. Then, either Save the changes or Save as a new Metric.

Additionally, if you need to duplicate an existing Metric, you can do so and save the duplicate to any Dashboard.