Contests

Company Name: Deloitte

Company URL: https://www.linkedin.com/company/deloitte/

Your Name: Jyoti Yadav

Your Title: Senior Consultant

 

Describe your company

We work as a consulting for a composite insurer.

From car, home and health insurance to pensions, investments and asset management, our client offers a wide range of insurance and saving products to help protect what's important to customer.

 

The customer experience and business challenge(s) you set out to solve with Adobe Experience Cloud products, and how long your company/organization has been using Adobe Experience Cloud products.

Customer Data is captured today through various channels but stored in silos. It takes up to a week to update customer data. Amount of data in use is not diverse, not leveraging all available sources. Marketing spends higher than any other industry​ and struggles to strategically scale investments.

 

Describe how you have integrated and used multiple Adobe Experience Cloud products to solve these challenges to improve and personalize the customer experience/journey.

We suggested to use the Adobe Experience Platform for 360 view of customer profile & leverage the functionality, Real-time CDP, to send stitched data to different destinations.

It will benefit in two ways:

1. Creating Real time Customer Profile providing single customer view – Unifying data from both online & offline
2. Sending real time customer data to different destinations

How we Integrated?

In layman’s term, client had below structure:

Now, using Adobe Experience Platform & Real Time CDP functionalities, changes in architecture is described below:

 

Stream or Batch data is ingested into AEP Data Lake from different data sources. Customer data is mapped using Identity and hence creating Identity Graph which creates unified profiles within the system. These Profile data is then used to build segments & send the Real-time CDP to different destinations.

 

With this set-up we were able to respond to customer behaviour in real-time by using a segment, on the condition that the segment qualifies in real-time.

 

Let’s understand the steps involved in achieving the result.

 

Step 1: We configure the required XDM schemas to classify profile information and customer behaviour. In every XDM schema, we must configure a primary identifier to link all the information.

From the business context we’re trying to link data to a customer & build a progressive, Real-time Customer Profile. We had two questions for the client to answer to categorise whether data belongs to the XDM Individual Profile or XDM ExperienceEvent Class.

1) Who is this customer? Answers to this will define data under Class XDM Individual Profile.

2) What does this customer do? We’ve defined in the class XDM ExperienceEvent - Experience Cloud ID (ECID) as the primary identifier.

Graphical representation of data classification:       

XDM Individual Profile			XDM ExperienceEvents
Field Groups			Field Groups
Demographic Details	Contact Information	Subscription Details	Web Details	Commerce Details	Environment Details
-First Name -Last Name -Address	-Email Address -Phone Number -Profile Picture	-Email Opt-in -Telecom Subscriptions	-Page details -Web interaction -Referrer	-Product Information: SKU, name -Cart Operations: order, checkout	-Device details -Browser info -Local time

   

After capturing above details, next step is to build an XDM Data Model.

 

Let’s understand the XDM Data Model design by considering 3 XDM schemas & how data is stitched among these 3 schemas, namely,

a) Recipient( to save customer demographic details),

b) Party( contains contact details associated with policy) &

c) WEB_SDK ( to save customer actions performed on website).

 

Create Recipient Schema:

1. Go to schemas, click on create new schema & select XDM Individual Profile.
2. Now define the schema name & select Field groups you want to be associated with this schema. We created one custom field group as well, to specify the Identifier. Next thing is to define the identifier which is key to Profile Stitching.
3. We had marked 'Master Id' as Identity in Master Schema.
4. The last thing to do here, is to activate the Schema to be linked to the Profile. So, activate the Profile toggle. This means you’re making sure that all data sent to Adobe Experience Platform against this schema will be part of the Real-time Customer Profile environment, which makes sure that all that data can be used in real-time for querying, segmentation and activation.

So, the 3 schemas have Identity defined as:

Schema	Identity
Recipient	Master ID
Party	Master ID &  MDMID
WEB_SDK	ECID & MDMID

 

Step 2:  Next step is to configure Data sets to create a bucket that uses XDM information, to receive and validate data that was sent to Adobe Experience Platform.

• Click on dataset -> Create dataset --> Select 'Create dataset from Schema' --> Define dataset name for both Profile & event schema--> Click the Profile toggle to enable this dataset for Profile.

After defining dataset, do the set-up to ‘Add Data’ from source Connection. An example is explained in below diagram.

• After scheduling the dataflow, we can validate the data ingestion by going to the path : Sources -->  Dataflows.

Step 3: Profile Stitching –

Over time, you can see the profile stitching across different sources. We were able to see identity overlap using Identity Graph.

    .     

How profiles are stitched over time.

Step 4:  Create Segments

Now data from different sources are not in Silos, they are all linked. i.e  All data that is ingested in Platform should be mapped against XDM, and as such, all data becomes part of the same data model regardless of where that data comes from.

 

We used data to create segments. Segments built within Segment Builder can be sent to solutions like Adobe Target, Adobe Campaign and Adobe Audience Manager for activation. We used these segments to sent to different destinations Google Customer Match, Facebook, Adobe Campaign. This can be achieved by creating segments & mapping segments to destinations.

 

Example scenario: Build a segment of all active customers who have purchased Healthcare Insurance.

Click on Create Segment --> Attributes - XDM Individual Profile --> Select browsing - drag & drop - add condition --> see an estimation of population count --> Save

Step 5: Configure an Advertising Destination like Google Customer Match

 

Go to Destinations --> Catalog( you will see the Destinations Catalog)  --> click on Google Customer Match(GCM) and then click + Configure --> Click data icon - Configure new destination( enter details & new destination is configured) --> Select Data Governance Policy ( it is optional)

 

Step 6: Send your segment to GCM - Map segment to Destination

Go to Destinations --> Catalog( you will see the Destinations Catalog)  --> click the Activate Segments on the Google Customer Match ( select destination & click next) --> select segment --> Schedule & finish

Your segment is now linked to Google Customer Match. Every time a customer qualifies for this segment, a signal will be sent to Google DV360 to include that customer in the Audience at Google DV360 side.

 

In graphical view, data in & out can be viewed as :

 

 

Based on your successful use and integration of multiple Adobe Experience Cloud products, describe how it has transformed the customer experience/journey, and the value, business impact, and results your company/organization has realized. Please cite both qualitative and quantitative results as applicable.

 

Quantitative:

• Stitched over 1M unknown to known profiles.
• We now have a reach of 1.2 million for display and 780,000 for search using only offline data

Qualitative Results:

• Merged Behavioural and Cyclops data could start to be leveraged as an input to segmentation used for activations in ways not possible before.
• With the CDP in place and all Behavioural events running through it, the Adobe Target OOB data connectors will be switched from the legacy Analytics to the new CDP datastream, i.e earlier it was Digital Analytics feeding data to target for Personalisation & now it will be CDP feeding data to target for Personalisation.
• Campaigns can be executed on CDP-managed audience segments for Digital Advertising, Messaging and Personalisation use cases.
• Cyclops data in the CDP resolved to a user can be used uploaded to GMP and other DSPs via native AEP connectors, which means current (semi)manual uploads from ACM into GMP and FB will be replaced, aka CDP à Digital Channels before it was ACM à Digital Channels.
• Customer Science models can be used in targeting for all channels.