Analytics Filtering

Kitewheel’s enhanced filtering capability supports deep journey exploration against a number of dimensions. These capabilities allow you to drill down by custom date ranges, channel, journey steps and journeys to gain insights into specific patterns in behavior and usage, rather than be hidden by noise. It can also highlight journey steps used in multiple journeys to help visualize the non-linear paths taken by your customers.

When applied, the filter settings impacts all the modules on the https://kitewheel.atlassian.net/wiki/spaces/K2/pages/697991173 accordingly. You can filter by:

  • Date

  • Channel

  • Journey Steps

  • Journey

By default, the Journey Discovery Analytics Dashboard will display data from the last full quarter of the year. All of the filters can be combined to drill-down to a very specific set of behavior. The filter affects all of the visualisations and the Dynamic Cards on the page.

Filter by Date

The date range filter is the global filter that will impact all the components on the https://kitewheel.atlassian.net/wiki/spaces/K2/pages/697991173. You can filter by date through the following ways:

Week

The Week filter selects seven days from the starting date you have clicked. E.g., Clicking on 11th of May 2020 would select the seven days data from Monday 11th May 00:00:00 AM through Sunday 17th May Sunday to 11:59:59pm Saturday inclusive. Each day starts and ends at midnight UTC.

Month

The example below shows a click anywhere in the month of June 2020 and will select from the 1st to the 30th of June 2020.

Quarter

The Quarter filter is split up into the four calendar quarters:

  • Q1: January, February and March

  • Q2: April, May and June

  • Q3: July, August and September

  • Q4: October, November and December

The default setting is for the last full quarter of the current year. For example in September 2020 the selection will default to Q2 (April, May and June) as that is the last complete quarter.

Year

The Year filter allows you to select a whole year, including the current year, all the way back to 1970.

Custom

The Custom date range filter allows you to input specific month, date and year. If the input dates are not valid, then the “Apply” filter button will not be activated and the date range text will remain as “Select Date Range”.

All date related information is stored and calculated in UTC.

Filter by Channel

The Channel filter allows you to select profiles that have had an interaction in the selected channel. This will be combined with the other filters and will impact the Activities Over Time, Most Popular Channels, Channel Behavior Flow and Interactions card on the Journey Discovery Analytics Dashboard and potentially the Total Profiles card as you are essentially filtering out interactions within a channel and Profiles tied to an interaction. By default, all channels are selected and you can remove channel(s) as needed.

Filter by Journey Steps

The Journey Step filter allows you to select profiles that have had one or more journey steps. The Journey Steps filter allows you to remove noise from the right and left Sankeys on the Journey Discovery Analytics Dashboard and focus in on key sets of journey steps. By default, all journey steps that have been created within the Hub would be included in the filter drop down and selected. You have the ability to search for specific journey steps to deselect as needed and you can select journey steps that have no data recorded against it if so desired. Search functionality will only allow you search within the drop down list.

The journey steps shown are as currently defined in the https://kitewheel.atlassian.net/wiki/spaces/K2/pages/748749236 page but may not match against the data exported into your Data Lake / Performance Dashboard. It is recommended to use the object ID (Journey Step / Interaction ID) to match appropriate step ID to the step name.

Filter by Journey

The Journey filter acts as a highlight to understand which journey steps are in which created journey map and highlight those in the Sankey. The information is pulled from your created Journey Maps and it is possible to select on a blank journey within the drop down. Each journey will be represented by a color within the Sankey. There will be five distinct colors (green, purple, orange, teal and red) to distinguish between journey one to five. The sixth journey and above will repeat those same colors.

If there are journey steps in multiple journeys, then the step node will display the various color separations on the node itself; however, hovering over the step node within the Sankey will pull up a tooltip as indication which journey(s) this step is in as shown below.

The Rule for Filter Combination

Journey Discovery Analytics Dashboard follows the inclusion over exclusion rule when it comes to counting the Profiles. This means removing a journey step from the filter will not necessarily remove that profile if they have taken an interaction in a selected channel and vice versa. As an example:

  • Person A did an interaction in the Web Channel

  • Then Person A performed a journey step ABC

  • Client filters out Web channel but keeps Journey Step ABC in the Sankey view

    • Effectively removing Person A from the Web Channel

    • But Person A will still be included in the Profile card because the client is including Journey Step ABC in the filter setting

To fully remove a profile:

  • Remove all journey steps that profile taken

  • Remove all channels on which they have interacted

  • Choose a time period in which they have no journey step or interaction recorded

Visualizations that will be impacted when using filters

 

Visualization

 

Visualization

Filter

Left / Right Sankey

Interaction Heat Map

Most Popular Steps

Channel Overlap

Date

Affected

Affected

Affected

Affected

Channel (Defined by Interactions)

Not Affected

Affected

Not Affected

Affected

Journey Step

Affected

Not Affected

Affected

Not Affected

Journey

Affected

Not Affected

Not Affected

Not Affected

 

Privacy Policy
© 2022 CSG International, Inc.