Overview

CleverTap's Trends feature provides advanced tools for analyzing and visualizing data, offering a comprehensive understanding of user behavior. The key functions include the following:

• Selection Capabilities: Users can choose event-level data, event properties, and user properties and specify analysis periods for deeper insights.
• Advanced Filtering and Segmentation: Robust filtering options and the ability to segment data for a targeted analysis.
• Visualization Tools: Trends can be visualized using various metrics, such as trendlines, conversion metrics, anomaly detection, and retention metrics.

These capabilities give users a powerful way to optimize platform usage and enable them to make informed decisions to drive business outcomes.

📘

Private Beta

Currently, this feature is a Private Beta Release. If you want access to this feature, contact your Account Manager.

Using the Trends dashboard, users can:

• Plot and analyze trends of event occurrences.
• Compare trends of multiple events to uncover patterns.
• Examine trends across different event property values.
• Compare event trends by user properties.
• Summarize trends using event properties.
• Compare trends across multiple segments.

Getting Started

The Trends dashboard enables you to visualize data by segmenting events based on event properties and user-defined segments. You can further refine these segments by splitting them based on user properties.

Results can be displayed in various formats, such as trendlines, pie charts, and more, and can be viewed over different time periods, such as days, or months. You also have the option to download the resulting visualizations or pin the analysis to a board for continuous updates. Additionally, you can download these results in a CSV file.

📘

Event and Segment Limit

You can select up to five events and five segments for analysis.

View a Trend

To set up a trend analysis, select the event you would want to analyze. You can also restrict your analysis further to a specific property of that event.

To view a trend:

1. Go to Analytics > Trends and select an event to analyze from the EVENTS section to view a trend. For example, you want to analyze all the users who launched the app.
2. Click View Analysis.

The trend results are displayed. Hovering over the trendline shows the percentage increase or decrease for the selected time frame and other event information.

You can change the view period of the trend on a daily or monthly basis:

• Daily: The trend and data points are plotted on a daily basis.
• Monthly: The trend and data points are plotted on a daily basis.

Event Count

To measure the event count:

1. Go to Analytics > Trends and select an event to analyze and view a trend.
2. From the MEASURE section, select any one of the following:
   • Total Events: This measure displays the sum of events in the specified period. If an event is performed multiple times, each occurrence is accounted for separately in the total count. For example, if user A performs an event on Day 1 and again on Day 2, the Total Events count will be cumulative, showing 2 for those two days combined.
   • Unique Users: This measure displays the trend for the number of unique users who performed the chosen events in the selected period. The Unique Users column in the table shows the unique users that have performed the event, property, and segment combination in the entire period selected from the date picker. For example, if user A has done the event on Day 1 and Day 2, the user is counted once each day. However, user A is counted only once in the Unique User column.
3. Click View Analysis.

View Event Count

To view the event count, hover over any trendline. For example, the number of App Launched events that occurred in the selected period. (This is repeated information - we have already stated earlier - hovering on the data point gives this information.)

The stats available are:

• Total events: Sum of events in the selected period.
• Daily change: The change in percentage from the previous time period, that is, day, or month.

Compare Trends for Events

You can compare trends for multiple events. For example, you can understand the correlation between two events, such as App Launched and Notification Viewed.

To view a trend, perform the following steps:

1. Go to Analytics > Trends and select an event to analyze and view a trend. You can also filter your analysis further by the event property.
2. Click Add Event to add another event for comparison.
3. Click View Analysis.

Filter by Event Property

When selecting an event for analysis, you can restrict the search criteria by filtering based on the specific event properties. All subsequent analyses are performed using these filtered event properties, while other properties of the selected event are excluded.

To filter a trend by event property:

1. Go to Analytics > Trends and select an event to analyze from the EVENTS section and view a trend. For example, you want to analyze only those users who viewed In-App notifications.
2. Click + Filter to filter by an event property that is In-App.
3. Click View Analysis.

The resulting trend filters users who viewed notifications from an in-app notification. It does not include users who viewed other types of notifications, such as Email or the Web. Any further analysis of this event will focus exclusively on users who viewed notifications from the In-App campaign.

Split by Event Property

You can split the analysis of your event based on specific event properties. For example, consider the Charged event, which indicates a user has made a purchase. You can split this event by its underlying properties, such as Payment Mode and Amount. This divides your single Charged trend line into multiple lines, with each trend line representing the value of a different event property. This allows you to visualize the payment method your users prefer.

To split a trend by event property:

1. Go to Analytics > Trends and select an event to analyze and view a trend. For example, Charged.
2. Click + Split by to divide the trend by the event properties, such as payment method.
3. Click View Analysis.

The trend for each event property is displayed.

All the split event properties are listed at the bottom of the right pane. You can perform the following actions:

• Use the search box to find specific event properties.
• Select or deselect event properties from the trend by selecting the corresponding boxes.
• Sort the event properties from highest to lowest values, or vice versa, by clicking the arrows.

📘

Maximum Split

You can split an event by up to two event properties at a time.

Compare Trends Between Time Periods

You can compare event trends against the previous period, such as day, week, month, and year.

To compare trends across time periods, perform the following steps:

1. Go to Analytics > Trends and select an event to analyze and view a trend. You can also filter your analysis further by the event property.
2. Select the date range for the trend. The trend displays.
3. From the Analysis pane, select the relevant time period from the Compare list.

A new dotted line appears for each series of the event data to display the difference in the values.

Compare Trends for Segments

You can compare your event trend across different user segments. For example, you can compare the App Uninstalled event trend across segments, such as Loyal Customers, Champion Users, and Match Day Players. Differences in the segment trends reveal variations in user behavior, providing invaluable insights. For instance, if many Booster Abandoners are uninstalling the app, it may indicate a need for alternative engagement strategies rather than offering them game boosters.

To compare trends across segments:

1. Go to Analytics > Trends and select an event to analyze and view a trend. You can also filter your analysis further by the event property.
2. Click Add Segment from the SEGMENT section.
3. Select a segment from the list or create an ad hoc segment. For more information, refer to Segments .
4. Click View Analysis.

📘

Comparison Considerations

You must consider the following when comparing trends across different segments:

• You can compare a saved or an ad-hoc segment. (Can we also compare a saved segment with an ad-hoc segment?)
• You can compare a maximum of five segments.
• You cannot split the trend by user property and compare it by segments simultaneously.

Adhoc Segments

When performing a trend analysis, you can use available user segments or create ad-hoc segments in run time. To create adhoc segments:

1. From the SEGMENT section, click the edit icon.

2. Click Add Rule to add a condition or click Add Rule Group to add multiple conditions.

For information about creating segments, refer to Segmentation.

Split by User Property

You can analyze a trend based on the specific user properties. For example, consider the event Charged, which indicates that the user has started watching content. Here, you can split this trend based on the underlying user property, such as Customer_Type.

This splits your single Charged trend line into multiple trend lines, each referencing the user property value. For example, the user property Customer_Type is split by user property values such as Platinum, Silver, and so on.

To split a trend by user property:

1. Go to Analytics > Trends and select an event to analyze and view a trend.
2. Select the user property from the Split by user property list under the SEGMENT section
3. Click View Analysis.

The trend for each user property displays. All the split user properties are listed at the bottom. You can do the following:

• Search through the listed user properties using the search box.
• Select or clear the user properties from the trend by selecting the specific checkbox.
• Sort the user properties from highest to lowest value and vice versa using the arrows.

📘

Maximum Split

You can split by a maximum of two user properties. Split by properties is not available when comparing two or more segments.