Funnels

Last updated:

A funnel is a behavioral reporting tool that allows you to measure how many visitors move through a defined series of steps in your app. Funnels can provide a high level of clarity as to where your visitors "drop off" when following these steps.

From there, you can use a path report to learn where visitors go from that drop-off point and better understand how visitors use your product. After you understand the path they take in your app, you can then leverage our guides to redirect your visitors through each step of the ideal route and improve the funnel completion rate.

Understand funnels

Funnels track a known sequence of steps through your app, like completing a specific task, to help you understand where visitors drop off in this flow. For example, if you want a visitor to start at your homepage, then select feature A, followed by feature B, a funnel can tell you if a visitor followed this route.

Funnel analysis options

There are two different ways to analyze funnels in Pendo: Unique Visitors and Total Attempts.

To understand which type of funnel analysis to choose, consider the main differences outlined in the table below.

  Unique Visitors Total Attempts
Recommended use case Understand whether or not users know how to complete a task Understand completion times for users who are familiar with a task
Questions answered

How many users are able to complete a specific task?

What's the conversion rate of users completing a specific task?

How long does it take users to complete a reoccurring task?

What's the completion rate of a specific task?

Timeout period Measured by Date Range selection (up to 90 days) Measured by Funnel Timeout field in the Analyze by section (60 minutes by default)
Multiple funnel entries Each completed step is only counted once. The funnel shows the first time the visitor generates each event within the defined order. If a visitor enters the funnel multiple times, each visit counts as a separate attempt. Revisiting the entry event resets a visitor to the beginning of the funnel.


If you're unsure which type of funnel analysis to use, consider running both options and comparing the results. While Unique Visitors can tell you whether a visitor completed a funnel and the conversion rate, Total Attempts can tell you the number of times a visitor entered the funnel and the completion rate.

How funnel completion is measured

Funnels ignore intermediate events between the defined steps. For example, imagine you have a series of funnel steps defined as A → B → C → D.

  • If a visitor takes steps A → B → E → C → F → D, this counts as a completed funnel as long as it's completed within the specified timeout period.
  • If a visitor performs A → B → C → E without performing D, they drop off after the third step as an incomplete funnel attempt.
  • If a visitor performs A → B → D without performing C, they drop off after the second step as an incomplete funnel attempt.

Let's also consider how a specific funnel could show different results depending on the type of analysis you choose.

  • On day 1, a visitor performs steps A and B.
  • On day 3, that visitor performs step C.
  • On day 5, that visitor performs steps C and D.

If analyzing the above scenario by Unique Visitors over a date range of 7 days, the funnel would show that it took this visitor 5 days to complete the funnel.

If analyzing the above scenario by Total Attempts with a timeout of 1,440 minutes (the equivalent of one day) in the same date range, the funnel would show that the visitor dropped out of the funnel after the second step since they didn't reach the third step within the defined timeout period.

Filter and group your funnel results

You can apply filters to your funnel steps or group your funnel data by common attributes using event filters and groupings. This allows for a more granular analysis of funnels, such as comparing funnels across different user groups or devices.

Funnels support filtering by event properties and historical metadata for each step, allowing you to refine usage data by specific properties captured at the time a visitor generated an event. Filter options include automatically-captured device data, metadata that's enabled as historical, and custom event properties configured for specific click events and Track Events.

Apply groupings to your funnels to break down funnel data by specific event property or metadata values. This helps you analyze conversion rates among different groups of users and identify which groups are completing the funnel at a higher or lower rate. Group by visitor metadata, account metadata, metadata that's enabled as historical, automatically-captured device data, and custom event properties configured for specific click events and Track Events.

When grouping by event properties or historical metadata, the grouping only applies to the first step of the funnel. This is because it isn't guaranteed that all events in the funnel have the same event property.

Tip: If you want to apply an event property or historical metadata value to every event in the funnel, consider using event property filters on individual steps instead of grouping the entire funnel analysis.

Create a funnel

  1. To create a new funnel, open the Behavior menu from the left-side menu, and then select Create report > Funnel.

    Behavior_CreateReport_Funnel.png

    Alternatively, you can navigate to Behavior > Funnels to view all existing funnels in your subscription, and then select Create funnel in the top-right corner of the page.
  2. In the first funnel step, select the appropriate event type and app from the top dropdown menus.
  3. Next, select the relevant event to be the starting point of your funnel. Each event type is described as follows:
    • Page. Displays all tagged Pages in the selected app.
    • Feature. Displays all tagged Features in the selected app.
    • Track Event. Displays all configured Track Events in the selected app.
    • Guide. Displays all guides in the selected app, where you can choose guides that were Seen, Dismissed, or part of the Not seen (control group) and a specific guide step. "Seen" indicates when a guide was launched to a visitor, regardless of what step it was launched on. "Dismissed" refers to when a guide was closed, either by interacting with a guide button or through any action that generates a guideDismissed event. The "not seen" option is only available if you have a control group experiment set up, allowing you to include visitors who haven't been exposed to the guide in your funnel analysis.

      Funnels_SelectEventType.png

  4. Repeat steps 2 and 3 for the second step in your funnel, then select Add funnel step for subsequent steps, up to 20 total steps. If you need to adjust the ordering of your funnel steps, drag and drop them to the correct placement.
  5. If you'd like to filter by an event property or historical metadata on any of the events, select the filter icon located to the right of the funnel step.
    • Select an event property from the dropdown menu that populates below the event, then select the correct operator. The default is equals to (=).
    • Select one or more property values from the dropdown menu. Note that any additional values selected for this specific filter are considered an “OR” rule.
    • To add another event property filter, select Add Filter, then repeat these steps for each new filter. Note that any new filters added to the event are considered an “AND” rule.
    • Repeat step 3 for each event you’d like to filter with specific event properties.
    • If you need to remove an event property filter, select the X icon located to the right of the filter.

      Funnel_EventPropertyFilter.png

  6. If needed, select a specific Date Range and Segment from the dropdown menus. By default, these values are Last 30 Days and Everyone. Custom date ranges are limited to 90 days.
  7. If you want to view the funnel separated by different metadata or event property groups, you can do so with the Group by section.
    • From the first dropdown menu, select the type of data you'd like to group your funnel analysis by, then select a metadata field or event property from the second dropdown menu.
    • When grouping by event properties or historical metadata, the grouping only applies to the first step of the funnel.
  8. Choose whether you want to analyze your data by Unique Visitors or Total Attempts.
  9. If you selected Total Attempts, update the default funnel timeout of 60 minutes if needed. For Unique Visitors, the date range is used as the timeout period.
  10. Select Save & Run to open the Save Report dialog.

    Funnel_SaveAndRun.png

  11. Enter a Report Name and Description so that you can easily identify the funnel report.
  12. If you want other users in your subscription to view and edit the funnel, update the Visibility to Everyone. (If your user account is only assigned the Viewer role, you can only choose Only Me.)
  13. If you’d like to add the report to an existing dashboard for quick access, select the dashboard name from the Add to Dashboard dropdown menu. On the dashboard, you can refresh the funnel data manually, or it automatically refreshes after each new report run and every 14 days.
  14. Select Save Report to create the funnel. You can now access the funnel through Behavior > Funnels or Behavior > View all reports.

    Funnel_SaveReport.png

Interpret your funnel

When you run a funnel, it can take up to several minutes to load the bar chart visualization and table breakdown. Each funnel component is explained in detail in the following sections.

Analyze the chart

At the top of the Funnel section, you can find three metrics:

  • Total conversion rate. This indicates the percentage of visitors who completed all of the funnel steps within the specified date range. It also notes the completed attempts (successful completions) and total attempts (which includes repeated attempts if analyzing by total attempts) from all visitors. This metric helps you gauge the overall effectiveness of the steps in your funnel.
  • Average completion time. This represents the average time it takes visitors to complete all of the steps within the funnel. When analyzing by unique visitors, this excludes visitors who didn't complete the funnel. This metric provides insight into the typical time for visitors to complete the funnel, considering all visitors' times, including potential outliers.
  • Median completion time. This reflects the middle value in the dataset of completion times for all visitors who successfully completed all steps within the funnel. When analyzing by unique visitors, this excludes visitors who didn't complete the funnel. This metric offers the typical completion time for most visitors in the funnel.

Even with a grouping applied, these metrics represent the overall conversion rate and average completion time for all groups within the funnel.

Funnel_Metrics.png

The funnel is represented by a bar chart located below the metrics, and each step added to the funnel is included in this chart. If you select multiple apps, the legend colors above the chart help you correlate your events with the relevant apps.

Funnel_Chart.png

Tip: You can easily understand the event type associated with a step by looking at the icon that precedes the event name on the X-axis. The folded paper indicates a Page, the mouse cursor indicates a Feature, and the lightning bolt indicates a Track Event.

The dark-colored area of each bar shows the number of visitors who continued to that step of the funnel, while the light-colored area shows the number of visitors who dropped off at that step.

Hover over the bar chart to view more information about a specific step:

  • The average and median amount of time it took visitors to go to the next step or come from the previous step.
  • The total number and percentage of visitors who started, continued, completed, or dropped off the funnel. For Total Attempts funnels, this includes all attempts from all visitors.

You also have the option to Watch replay samples if you have access to Session Replay. For more information, see Watch replays.

Funnel_StepHoverTooltip.png

Tip: Use < Tab > on your keyboard to activate this hover state and the arrow keys to navigate between steps.

Below is the same example chart as earlier with a grouping applied. Groups are ordered in descending order based on the number of visitors who completed the funnel.

Funnel_ChartGroups.png

If you have more groups than can fit on the page, you can select Next in the top-right corner of the chart, scroll through them in the chart using a trackpad, or use < Tab > and the arrow keys on your keyboard. If using your keyboard, you can use < Enter > to select a group and then use the arrow keys and < Enter > again to select a specific step. To reset the group selection, use < Shift + Tab > then < Enter > when All Groups is highlighted.

Select a group in the chart to view the percentage of visitors in that group who completed the funnel, the average and median time it took visitors to complete the funnel, and the total number and percentage of visitors who started and completed the funnel. For Total Attempts funnels, this includes all attempts from all visitors.

Funnel_GroupChartHover.png

After you select a group in the funnel chart, you can select a specific funnel step to view metrics specific to that step (the same tooltip information that shows when you're not grouping your funnel).

Tip: If you see lower conversion rates in specific user groups within a key funnel, consider using a Pendo guide to improve conversions among those groups.

View the breakdown

The Funnel Breakdown section contains a table that summarizes the data shown in the chart. There are three main table views that vary depending on your selection.

The first view that shows by default is the Funnel Breakdown Summary. In this view, the table includes the total conversion rate of users, the total number of unique visitors or total attempts by all visitors (depending on your selected analysis option), the total number of visitors or attempts completed, the average completion time, and the median completion time amongst all visitors. With a group applied, the table shows a summary for each group.

Funnel_BreakdownSummary.png

The second view is the Funnel Breakdown by Step, which you can view by selecting the Summary dropdown in the table title and selecting Step. When grouping, the table groups all steps together so that you can view the conversion rates for each group.

In this view, the table breaks down each row by step and provides the total number of unique visitors or total attempts by all visitors (depending on your selected analysis option), the conversion rate for each step, and the average time it took all visitors to continue to each step from the previous step. Any applied event property filters display in the Filters table column.

If you have a grouping applied, you can select a group from the chart to view the breakdown for a specific group. Any applied event filters display in parentheses next to the event name.

Funnel_BreakdownByStep.png

The third view is a Step Breakdown, which you can access by selecting a step in the bar chart. When selected, the step in the chart is highlighted a darker color and the event name populates in the table title with any applied filters in parentheses.

  • If you select the dark-colored portion of the step, the table shows data for Visitors who continued, which lists the visitors who continued to that particular step, the date and time each visitor reached that funnel step, and the time it took them to proceed to the next step. If a visitor didn't make it to this step of the funnel, their Time to Next Step value shows as --.
  • If you select the light-colored portion of the step, the table shows data for Visitors who dropped, which lists the visitors who dropped off at that step of the funnel.

To change your view from Visitors who continued or Visitors who dropped, select a different part of the bar chart or use the dropdown menu in the table header. The number in the table header corresponds with this selection. For example, in the following image, 1,042 denotes the total number of visitors who continued to the third step in the funnel. If you have access to Session Replay, you also have the option to Watch replay samples for these visitors. For more information, see Watch replays.

Funnel_StepBreakdown.png

This table displays the Account ID associated with each visitor. For visitors who dropped off, the Account ID value corresponds to their last step in the funnel.

To exit the Step Breakdown view and return to the Funnel Breakdown Summary, select Reset view or All Groups (depending on if you have a grouping applied), or unselect the selected step in the bar chart.

Note: The table view is limited to 10,000 rows per step. However, you might see more than 10,000 visitors because visitors who continued to the next step don't count toward the limit of the previous step. If you need to view visitor data for every event in the funnel, consider adjusting your segment to a smaller cohort of visitors. You could also consider updating your steps to follow a more specific route and avoid primary navigation buttons.

If you want to export the breakdown view as a CSV file, select the Download CSV icon located in the top-right corner of the table section. This export option includes the data that's currently shown in your table.

Funnel_Breakdown_DownloadCSV.png

If you want to instead export a list of all visitors who entered the funnel, regardless of completion, use the Download CSV icon located at the top-right of the page. This export option includes the Visitor ID, number of steps each visitor completed, and relevant timestamps for each event.

Funnel_DownloadCSV.png

Share your funnel

As long as you choose the Everyone visibility at the time you save the funnel, all Pendo users in your subscription can access the report through Behavior > Saved Reports or Behavior > View all reports.

If you need to update the visibility, simply select your report from Behavior > Saved Reports, ensure Visibility is set to Everyone using the Share icon located at the top of the report, then share the URL found in the address bar in your web browser or let them know it's available to view on the Saved Reports page.

Funnel_Share.png

As an alternative to direct sharing, you can also add the report to a dashboard using the Funnel widget and share the dashboard with that user. You can do this by adding a new dashboard widget to a dashboard or by selecting Add to Dashboard at the top of an existing funnel report or when you save a new funnel report. On the dashboard, you can refresh the funnel data manually, or it automatically refreshes after each new report run and every 14 days.

Funnel_AddToDashboard.png

Was this article helpful?
7 out of 10 found this helpful