This blog details a single implementation or customization on Adobe Experience Platform. Not all aspects are guaranteed as general availability. If you need professional guidance on how to proceed, then please reach out to Adobe Consulting Services on this topic.
We have been experiencing global new trends in customer experience since early 2020. Retailers, in particular, have had to adapt to new safety and security public health requirements while committed to delivering exceptional customer experience.
For example, real-time offers from anywhere are imperative to offer the customers the best value as well as delightful interactions. Web, mobile, and curbside shopping must be seamless.
For CIOs and CMOs, actionable reporting and real-time data science insights have become mission-critical to deliver these new and improved customer experiences in these extraordinary times.
The customer highlighted in this blog had reporting needs for online and offline marketing. The data that retailers need to correlate revenue per household was in silos. Thus we needed a better understanding of their marketing spend and ROI.
For marketing campaigns, how do you get the right attribution so the retailer can improve and plan for the next campaign?
It’s about the ability to activate omnichannel for campaigns at scale and understand the behavior at the household level. Furthermore, the need was there to target and personalize based on the individual and household.
Operationalizing the data science models for near-real-time personalization
Two examples of use cases were:
If a person is doing fridge online and purchases fridge offline at the store, merge the data to deliver the next best experience in real-time.
If a person is browsing on the website and delivers the next best action recommendation while they are browsing the website.
Customer Use Cases and Desired Outcomes
The POC was focused on three major business priorities for our customer:
Figure 1: Customer POC Focus Areas
Customer Journey Insights: To build the reporting for the online and offline transactions and build campaign performance analysis and help our customers understand the customer behaviors and purchase patterns online and offline. Ability to build the campaign performance analysis by aggregating the data at the campaign level across multiple sources and channels.
Connected Experiences: Provide the ability to build customer profiles, segments, and the ability to activate on multiple channels. The major business goal was to onboard the various identities in the customer ecosystem and be able to build the profile and then make the profiles actionable by building segments and further pushing the segments on multiple channels.
Data Intelligence: Leverage the AI/ML technique to provide and score models at runtime. Ability to use the Data Science Workspace to build models using the customer data. Also to prove the ability of BYOM Bring Your Own Model and operationalizing them can be accomplished to provide the real-time aspect.
The POC goals and objectives to overcome the current state challenges:
Figure 2: Customer POC Use Cases Summary and Goals
Adobe product in the play:
Adobe Experience Platform
Real-Time CDP and Customer Journey Analytics
Figure 3: Customer POC Architecture
The enterprise architecture for end-to-end execution.
At the core, the architecture demonstrated the onboarding of data into Adobe Experience Platform via Streaming, batch connectors, and then augmenting the data into Adobe Experience Platform by building profiles, segments and orchestrating it for Activation and reporting.
The data onboarding into Adobe Experience Platform leveraging the OOTB connectors and by developing the XDM schemas and data sets in Adobe Experience Platform.
As part of the POC, we developed 12 + custom schemas and onboarded data using the OOTB Adobe Analytics connectors (Streaming), GCS connectors as batch-based ingestions. Adobe Experience Platform offers many out-of-the-box source connectors to ingest the data. We also used XDM to define schemas, mixins, and identities for the customer data needs. This allowed persisting the data in a cohesive manner for larger adoption.
We successfully onboarded the customer identities to be able to build a holistic profile and allow for activation across channels.
In addition, we also leveraged Query Service as well to perform various data aggregations to generate insights.
Building the Real-Time Customer Profile
Building the profile in Adobe Experience Platform, Segments, Setting up destination via Real-time CDP, and propagating the data to Customer Journey Analytics for reporting.
Once the data is on-boarded, we can mark data sets and schemas for a profile that allows the Adobe Experience Platform to build the customer profile.
Ansible UI allows for segment creation capabilities
We can leverage the Real-Time CDP destinations to set up the connections
All the data onboarded in Adobe Experience Platform can be propagated to Customer Journey Analytics by building a connection in Customer Journey Analytics and using a common id across the data sets.
Leverage Real-Time CDP for Activation
Leveraging the Real-Time CDP for activation across multiple channels and using the Customer Journey Analytics for the reporting. As part of the activation use cases, we used the Adobe Experience Platform destinations to integrate with the following destinations
Facebook: Use the hashed email address as the identifier.
Google customer match: Use the hashed email address as the identifier.