Journey Visualisation View

As events flow into the system, they are created into Event Streams (Queue) to ensure that they are then stored in the correct order. Typically there are three Event Streams available, corresponding to the three Targets in the Journey Orchestration., which are Dev, Test and Prod.

The Journey Visualisation represents these as behaviour journeys for each Visitor, then looks for common behaviours across all Visitors in the same date range and displays the top N of those in the Journey Visualisation.

After the initial Homepage is displayed, giving access to any saved Visualisations, a saved document can be opened, or a new Journey Visualisation can be created clicking the + icon

When opened a blank Journey screen will be displayed until a selection of date range and other required parameters is carried out. An auto-calculate toggle can be set to allow saved Journey Visualisations to calculate on opening. The default configuration for new Journey Visualisations is the “Dev” Event Stream activity over the last 7 days displaying by Channel + Location with no filters.

This main areas of the Journey Visualisation screen are defined below.

Open

Journey Visualisation Flow

The main area of the screen is the Journey Visualisation itself. This shows the transition of Visitors between the various Locations (interaction points) that have occurred over the defined Date Range and within the scope of any defined filter, with a colour coded view of Channel. The bigger the node on the screen or the bigger the connection flows, the more Visitors have passed through this location or followed this flow. The red drop-off indicator shows the end of a stream or Journey.

“Other Event” nodes represent the Locations that are not in the top N (Depth Setting - see later), so this is an aggregated node of Location + Channel events that are not displayed specifically and are not representative in the data.

“No Event” nodes appear when performing a right alignment (more later), where actually No Events exist prior to the streams being right aligned or in other words, no events have occurred in a 30 minute period to the left of that current Location.

File and View

For the purposes of this screen a “file” is a single Journey Visualisation.

The File menu allows the Visualisation to be saved, allows the definition of a name and a Public Access (everyone) or Private toggle. If this option is not selected, this view will only be available to the creating user on the Journey Analytics Homepage. For previously saved Visualisation. The File menu also allows the option to Make a Copy, and using the same dialogue as before, to save this as a new file. The file menu is then also used to close the current visualisation.

View menu, provides some very powerful and useful features for displaying multiple Visualisations. Having two Journey Visualisations side by side allows quick comparison, perhaps between dates ranges or any filtered value. A second or subsequent Visualisation can be added before (to the left) or After (to the right), either creating New, Adding a duplicate or Loading a previously saved file. Each visualisation is it's own File and can be interaction with as a separate entity.

Copy setting from allows a currently open File’s configuration to be imported into the selected file and each visualisation can be removed from this view.

Event Stream Queues and Keys

This option allows the selection of an Event Stream Queue and a Key.

• Event Stream Queue - Each CX “Client” will have a backend Event Stream associated with it to collect data from the Visitor interactions and load them into the database for analysis here. Each Event Stream typically has three “Targets”, being Dev, Test and Prod, which in turn allow the analysis of Visitor events generated in the target as specific by the rule in Rule Builder. The name of the Event Stream is a concatenation of the defining pieces, for ease of use.

• Key - The system has two “keys” known as Visitor and Interaction. This is a very important subject, and more information if available on the Visitor and Conversation Keys reference page. Once understood, this selector allows the switch between the two Key views giving more flexibility for data analytics. At a high level, for Visitor, the stream being analysed is all events for a single Visitor, through all Interactions, based on the filtering date range etc, whereas the Interaction Key treats each Interaction as a different stream for analytics across a single interaction flow.

Date Range

Selecting the Date range is a very important factor when wishing to create a Journey Visualisation. The Date range covers the dates between which the data will be collected, converted into specific Journeys (Visitor Key) or into Interactions (Interaction Key), for subsequent analytics. Once a Date Range is selected and executed, all subsequent Attribute Filtering queries are performed against the data from that date range.

IMPORTANT NOTE: Each time the date range is changed, and new set of data is being submitted for Stream processing, and this process can take some time to perform. Date ranges should be chosen carefully as each change does incur significant processing time.

The following Date Range filter options are available, noting that all dates work from 00:00:00 to 23:59:50 of the defined days.

Selectable from UI defaults

• Last 7 days - going back 7 days including the current day working back.

• Last 30 days - going back 30 days including the current day working back.

• Last 60 days - going back 60 days including the current day working back.

• Last 90 days - going back 90 days including the current day working back.

• Last 365 days - going back 90 days including the current day working back.

Custom Dates

To define a custom date range simply select the start and end dates from the calendar screen to define the scope of the query.

Date Range Data Validation

• If you are approaching the Journey Visualisation for an initial view, we have defaulted to the last 7 days. This is the lowest of the pre-defined values and is chosen based on comments above regarding the time to perform the initial data processing.

• If you are approaching the Journey Visualisation to see activity from a known Location, then you will want to ensure that it is representative enough in the data within your Date range. Having selected a Date Range, it can be useful to use the Align by option to get a direct view of the number of times the Location occurs in the selected Date Range. It maybe be, that the Location you are interested is not visible on the default Journey Visualisation, as it is not prominent enough. The Depth and Width parameters are covered later, but suffice to say that the initial Journey Visualisation view is only a subset of the Journeys or Locations that exists in the data behind your selected Date Range. If the desired Location is sparse, then an increase of Date range can ad more data to be viewed, but again this will take some time for processing.

Group By

Group By is a very powerful tool allowing the boxes or nodes on the Journey Visualisation, to represent different entities in the system.

The default view in a Group By is Location and Channel, which can be seen in the screenshot at the top of the page. In this view, each node represents a specific Location and Channel combination that the Visitor has engaged with. For example with a Banking website, these might be Homepage, Personal Accounts, Mortgage Overview etc, with the colour coding displaying there Channel of interaction.

Also, as part of implementation and on-boarding, the Location values on the system will be refined to ensure they are shorter and more concise than the full URL. At this stage, it may be that two Locations of the same name exist on two different Channels, for example both the Website and the App might have a /Homepage, and for this default view, there would be one node for Homepage on the App, and one for Homepage on the Web as seen below.

Open

Another way to approach the analytics is to group the Locations into more generic Categories. For example with Banking again, any Locations that relate to Personal Account Management could be put in the “Personal BAU” Category. Selecting the Category as the Group By, switching to a Category and Channel Group By, would change the Visualisation nodes from Locations to Categories, showing how a Visitor is moving between Categories and Channel, rather than Locations.

The Group By options are summarised as follows:

• Location - In this view each box represents a Location with no Channel differentiation. In the example of having two Homepage Locations above, one for App and one for Web, these would now be considered the same. The node on the Journey Visualisation for Homepage would include both App and Web.

• Location and journey - This option is available if the Location Mapping & Journey Exploration feature has been used and add a Journey Label to the UI for those Locations associated with a Journey via Location Mapping

• Location and Channel - In this view each box represents a Location and Channel combination. In the example of having two Homepage Locations above, one for App and one for Web, this would would consider them as separate nodes.

• Category - In this view each box represents a Category of Locations with no Channel differentiation. This shows Visitor behaviour between Categories without differentiating by Channel. (note depending on implementation, Category may not initially be populated.)

• Category and Channel - In this view each box represents a Category of Locations combined with the Channel. This shows Visitor behaviour between Categories with Channel differentiation. (note depending on implementation, Category may not initially be populated.)

• Channel - A very high level view of how Visitors interact between Channels.

• Journey - This option is available if the Location Mapping & Journey Exploration feature has been used and used each node in the Journey Visualisation to be associated with the Journey that Location represents. When Show Unique Nodes is True, each node may represent multiple Locations consecutively from the same Journey, so will simply show transitions between Journeys.

When viewing a Group By, there is also the option to Show Unique Nodes. This is a powerful tool for analysts as it provides a useful “dedupe” option. Imagine a scenario where a Visitor to the Brand has trouble with a “password Reset” The Journey Visualisation would see repeated Nodes of “Password Reset”, and this can be very useful when looking for candidates for some re-work orchestration to improve the Customer experience. On the Journey Visualisation, these repeated patterns would be seen as repeated nodes on the flow. Once identified the Analyst would then want to see what came before and after those repeated patterns. The Show Unique Nodes option, when set to True, would represent those repeated Password Resets as a single node to help the analyst focus on behaviour before and after.

Show Unique Nodes

This option, set to True by default, allows the grouping of consecutive events for the same Location. If, for example, you see three add to basket Locations in a row, it might be the user had to refresh or was perhaps re-entering some information that refreshed the page, this pattern would appear. The True option here would only show this as one Node, where as False would display as three Nodes.

Attribute Filters

This area is provided to further filtering of rows

Display

• Align - allows the Journey Visualisation to be right or left aligned giving great analysis at either the start or end of the stream of activity in the defined filter criteria and key.

• Align by - Using the alignment option, further focus can be added to the visualisation by right or left alignment on a specific event location. Clicking on this option disalys a pop up looking at all the data in the Date range, and supplies useful counts on the number of occurrences of each Location, or other Group by. The Journey Visualisation view analysis gives a very useful overall view of streams of activity, but it is important to understand that each vertical position is a step in the journey and the default view takes that into consideration. If, for example, analysing steps leading up to a Drop Basket location event, for some visitors this might be the 5th interaction step in a Interaction stream, and for some, the 10th, so in the default view, these would be different Location positions. By using this option, specifying a “Drop Basket” location right aligned, this would “group” all instances of a Drop Basket in a single Location on the right of the screen and show all streams leading up to that event, regardless of how many location steps lead up to the specified event. This is very powerful when wanting to analyse stream patterns to a specific action performed by the Visitor. The same functionality can also be achieved by clicking on a node to display a menu to that that object as either right or left aligned.

• Width - used to adjust the number of Locations displayed horizontally, and has a default of 5 and is simply displaying more Locations in the stream. This option has a maximum value of 20 as it is felt beyond that point the visualisation is not useful. After the defined number of Locations, the remaining are aggregated together and displayed in a single Location object called “Other Events”.

• Depth - is used to define how many “common” streams are to be displayed. As part of the Journey Analytics calculation, the streams are all analysed and compared so the diagram displays the most common streams and the depth setting described how many of those common streams that Analyst wishes to see. This option has a maximum value of 20 as it is felt beyond that point the visualisation is not useful. An “Other Event” node is visible vertically in each Stream position to represent an aggregation of other Locations.

Other Visualisations

As well as the flow visualisation this screen offers two other useful visualisations. They display based on the same filter calculation as the Flow diagram.

Visitor Stats

This screen shows some high level stats of Visitor and Visitor Interactions that occur during the specified time period and based on the filtered data

The following table shows details of the stats available in this screen.

Relationship

A flow based visualisation at the Location level, showing the relationship and flow between the Locations returned within the defined date range and respecting any attribute filters. The size of the outer bar showing creative numbers of Visitors for each Location and inside the circle, the flow between one and the next. It should be noted the visualisation looks at the Location, regardless of its position in a Journey or Behaviour path.

Distribution

A Bar Chart distribution of the Locations showing count of Locations returned within the defined date range and respecting any attribute filters. It should be noted the visualisation looks at the Location, regardless of its position in a Journey or Behaviour path.

Journey Stats

This feature is dependant on the use of Location Mapping & Journey Exploration, to define and label the Locations relating to the Journeys present in the data, as well as ensuring the correct and full data is being collected for those stats within the rules in Rule Designer.

There are three standard metrics used to create this Journey Stats view.

• Conversion Value - Configured in the Conversion Event, the Conversion Value is populated and captured in a single Conversion Value column, which is them available for aggregation to a Journey Level.

• Service Cost - Configured on any event Type to capture a standard Service Cost against the interaction point.

• CSat - A standard derived CSat score from Satisfaction Values captures in Satisfaction Event Type

The statistics are calculated as seen below