Adobe CJA Business Practitioner - Components

Connection

Data View

Information available in the data dictionary

A component's function and intended use, components typically used with the one you're viewing, components similar to current viewing, whether component is approved by system admin

Benefits of data dictionary for admin

Identify duplicate components, identify components that aren't collecting data, identify components that aren't approved

Data Dictionary

allows you to view information about a component, including the component description, similar components, other components a component is frequently used with

Type of components in data dictionary

Dimensions, Filters, Data Ranges, Metrics

Components within the data dictionary are one of two approval statuses

Aproved by administrator, or Unapproved

Key filter abilities in Data Dictionary

Show duplicates, Missing Description, No recent data, Created by adobe

Key sorting abilities in Data Dictionary

Recommended (components used most frequently by you and others in org), Last Modified, Alphabetical, Categorical

When you select a component in Data Dictionary, what information is displayed

Approved (admin sees option to unapprove), Not approved (admin see an option to approve), Description, Frequently used with, Similar to, Product compatibility, Tags, Component Type, Created By, Preview, Date Last Modified

"Frequently Used with" in Data dictionary

Shows components most commonly used with component you are viewing. Up to 5 components across the 5 component types are displayed. List is based on past 90 days

What can Admin curate in regards to "frequently used with" and "similar to" in Data dictionary

Always Include, Always Exclude drop downs

"Similar to" in Data dictionary

Shows components with similar names. Up to 5 components across 5 component types. Only components you have access to are shown. Any duplicate components will display.

"Product Compatibility" in Data Dictionary

Indicates where in CJA this calculated metric (component) can be used. Options are Everywhere in CJA or Everywhere in CJA (excluding experimentation)

Annotations

enable you to communicate data insights by letting you tie calendar events to dimensions/metrics. You can annotate a date range with known data issues, public holidays, launches. You can then graphically display these events

Annotations example:

Annotation Bad Offers for specific date range where offers being accepted went down significantly. Bad offers is scoped to your whole data view. What is the result?

Any dataset that includes this date range will include the annotation within their projects, alongside their data.

Annotations can apply to

A single date or date range, Your entire dataset (or specific metrics, dimensions, or filters), the project in which annotations are created (default), all projects, the data view in which annotations are created (default), or all data views

What can you use to manage annotations?

Annotations manager

At what levels can you turn annotations off?

Visualization, Project, User

Bounce Rate

Calculated Metric. The percentage of website visitors who view one page and then leave the site

How do you define a calculated metric Bounce Rate in CJA

devine a Bounced events filter where Session Start equals 1 and Sessions end equals 1. You can also define using derived fields

What are derived fields?

Derived fields are part of a dataview and allow you define data manipulations on the fly through customizable rule builder

Conditional Page Views

Calculated Metric that calculates on page views that have been visited based on X (visitied for > 100 seconds)

Metrics

Allow you to quantify data points in Analysis Workspace. Most commonly used as columns in a visualization and tied to dimensions

Dragging a metric on top of an existing metric header ...

replaces it

Dragging a metric next to a header lets you ...

see both metrics side by side

Standard metrics types

People, Sessions, Events

Standard metrics differences (CJA VS AA)

CJA allows you to define standard metrics in a flexible way, within the scope of a connection and data view

Standard metrics

People

The count of distinct PersonIDs (depending on what you choose as the Person ID when you configure datasets in your connection, the People metric can mean different things)

Standard Metrics

Sessions

What you define as part of the configuration of your data view.

Standard Metrics

Events

Comprised of the events that are part of any dataset you have configured as part of your connection

Calculated Metrics

User-defined metrics that are based on standard metrics, static numbers, or algorithmic functions

Calculated metric templates

Adobe-defined metrics that behave similarly to calculated metrics (you can use them out of the box, or copy them to edit )

Calculated metric builder

used to create new or edit existing calculated metrics

What can you specify in calculated metric builder

Data view (metric you define is available in workspace projects based on data view), Project-only metric

Definition Builder (Calculated metric builder)

drag and drop dimensions, metrics, filters, and functions to create custom metrics

CJA Audience Publishing

you can create and publish audiences discovered in CJA to AEP

What can you do with audiences discovered in CJA

Use an audience in AJO, exporting to a third party through AEP destination

Do the Audiences you create in CJA have to be based on datasets enabled for profile?

No

Audience

A set or list of identities that have both a namespace and specific ID related to that namespace. Audiences can contain mixed namespaces

Filter

A set of rules that, when evaluated over a set of data for a time period, produces a subset of data. Filters are defined and maintained in CJA

What is the purpose of a filter in data analysis?

A filter is used to narrow down a dataset for analysis purposes.

What is a segment used for?

A segment is used to produce a list of identities that can be used for activation.

Does CJA support the concept of segments?

No, it uses filters. While both are a set of rules that contain similar logic, they produce different outputs.

What do segments produce in Real-time Customer Profile?

Audiences

Do filters alone produce audiences in Real-time Customer Profile?

No

What is CJA audience publishing?

The process of using CJA filters to create audiences for Real-time Customer Profile

Participation Metrics

Used to quantify how individual values for dimension (like page views) contribute to, or participate in sessions that contain specific metrics

What can filters be based on

attributes, interactions, exits and entries, custom variables

Standard Filter Operators

filter based on standard operators (equals, etc)

Distinct Count filter Operators

You can filter based on a count of items within a dimension

Quick Filters

Allow you to explore data within a given project, without the need of creating more complex filters in the filter builder

Stitching

Identity stitching (or simply, stitching) is a powerful feature that elevates an EVENT dataset's suitbility for cross-channel analysis

Cross Channel Analysiis

Ability of CJA to combine and run reports seamlessly on multiple datasets from different channels based on a common identifier (person ID)

Are all event datasets eligible for

CJA stitching

No. Some event datasets are not sufficiently populated for this attribute out of the box (especially web-based or mobile-based experience datasets don't have actual person UD information on all events)

Stitching authentication

Looks at user data from both authenticated and unauthenticated sessions

Stitching limitation - Persistent ID

you cannot use identityMap as persist Id, you have to use a specific identifier in the dataset (e.g. ECID)

Are all dataset types permitted for stitching

No, only event datasets

Is the stitching process case sensitive?

yes

Persistence

Persistence is the ability for a given dimension value to relate to a metric beyond the event it is set on. It uses a combination of allocation and expiration

Standard component reference

components that are not generated from dataset schema fields, but are instead system generated.

Bar and bar stacked visualization

shows vertical bars representing various values accross one or more metricss

Are audiences immediately published from CJA to AEP realtime customer profile?

No, there is a series of steps.

What happens if you accidentally delete an audience you've created in CJA?

It persists in AEP until the profile membership of the audience in AEP expires

Does the Audience in AEP (created from CJA filter) share the same name and description as the CJA audience?

yes

What can you say about the name of an audience created in CJA and published to AEP

the name of the audience in AEP will have the audienceID appended to it

If you make a change to an audience in CJA, will it reflect in the audience in AEP

Yes

What happens if a user is no longer a member of an audience in Customer Journey Analytics?

In this case, an exit event is sent to Experience Platform from Customer Journey Analytics.

If a corresponding profile does not exist in Real-Time Customer Data Platform, is a new profile created?

Yes

Does Customer Journey Analytics send the audience data over as pipeline events or as a flat file that also goes to the data lake?

Customer Journey Analytics streams the data into real-time customer data platform via pipeline, and this data is also collected into a system dataset in the data lake

What identities does customer journey analytics send over when publishing an audience to AEP?

Whichever identity/namespace pairs were specified in the Connection setup. Specifically, the PersonID field

The PersonID of a connection setup becomes WHAT when an audience is published from CJA to AEP

the primary identity

Attribution Model

An attribution model determines which dimension items get credit for a metric when multiple values are seen within a metric's lookback window.

Last touch attribution

gives 100% of the credit to the touch point occurring most recently before conversion

Last touch attribution: typically used when ...

Default for any metric where no attribution model is specified. Use when time to conversion is short (analyzing internal search keywords)

First Touch attribution

Gives 100% of the credit to the touch point first seen within lookback window

First touch attribution: typically used when ...

Organizations want to understand brand awareness or customer acquissition

Linear attribution

Gives equal credit to every touch point leading up to conversion

Linear attribution : typically used when ...

conversion cycles are longer or require more frequent customer engagement (typically used for measuring mobile app notifs or with subscription based products)

Participation attribution

gives 100% credit to all unique touch points (metric data typically adds up to more than 100%)

Participation attribution : typically used when ...

organization wants to understand which touch points customers are exposed to the most (content velocity)

Same touch attribution

gives 100% credit to the same event where the conversion occurred (this attribution model is sometimes equated to having no attribution model at all)

U shaped attribution

Gives 40% credit to the first interaction, 40% credit to the last interaction, divides the remaining 20% to any touch points in between (single touch point gets 100%, two touch points each get 50%)

U shaped attribution model: typically used when ...

you value first and last interactions the most

J curve attribution model

gives 60% credit to the last interaction, 20% credit to the first interaction, divides the remaining 20% to any touchpoints in between

Inverse J attribution model

Gives 60% credit to the first touch point, 20% credit to the last touch point, divides the remaining 20% to any touch points in between.

Time decay attribution model

Follows an exponential decay with custom half-life parameter, where the default is 7 days

Custom attribution model

allows you to specify the weights you want to give to touch points

Algorithmic attribution model

Uses statistical techniques to dynamically determine the optimal allocation of credit for the selected metric

Lookback window

amount of time a conversion should look back to include touch points

Session settings

in CJA you can define a session in any way to match how persons interact with your digital experiences

Use a short session timeout when (eg 30 mins)....

you are analyzing mostly online interactions (whether profiles visiting online store or product pages added to their cart)

Use a long session timeout when (eg 3 months)...

you are combining online and offline data and want to analyze multiple events/channel interactions

Person Lookback window (Reporting Window)

Looks at all visits back up to the first of the month of the current date range

Persistence is based on

Allocation and Expiration

Allocation (persistence)

lets you determine which value is kept when more than one dimension item can persist at a time in a single column

What do you set attribution models on

components

Expiration (persistence)

Lets you determine how long a dimension item persists beyond the event it is set on

is persistence retroactive

Yes

Persistence is only available on

dimensions