If you’re getting ready for a Salesforce Data Cloud implementation, then this post will get you ready for it.
There are so many buzzwords with this particular Salesforce product that it often makes it hard to understand what Data Cloud is and what it can do for your business. If that sounds familiar, this article should help you understand the product’s core capabilities and key considerations.
Having been lucky enough to work on a Data Cloud implementation, I’ll be drawing on both my theoretical knowledge from countless Trailheads and accreditation courses as well as my practical understanding from the challenges I faced when implementing this intricate cloud.
What is Salesforce Data Cloud?
First things first, Data Cloud, like a lot of Salesforce marketing products, has been through a lot of rebranding and was formerly known as CDP. However, it is not to be confused with Salesforce 360, which will leverage the power of Data Cloud across the whole Salesforce Ecosystem.
So what actually is Data Cloud?
Simply put, it is Salesforce’s long-term customer data platform (CDP) solution. The platform allows users to create a unified view of their customers by integrating data from multiple sources, both internal and external. This data can include demographic and behavioral information, purchase and order history, digital and non-digital interactions, and much more. By combining these diverse datasets, users can gain a deeper understanding of their customer’s preferences, behaviors and most importantly, needs.
Source: Salesforce
What are Data Cloud’s Capabilities?
If you’re looking at implementing Data Cloud and wondering whether or not it is right for you, let me talk you through its capabilities in a little more detail:
Data Unification
First and foremost, Data Cloud’s primary function is to unify data from multiple sources into a single, consolidated view of the customer. Data can come directly from Salesforce Core, Salesforce Marketing Cloud or it can come from an external system such as an in-house data warehouse. However, as I’ll go on to explain later on, Data Cloud works best when your data is in a healthy position.
Data Enrichment
With a plethora of connection options ranging from out-of-the-box (OOTB) connectors to FTPs to APIs and more, Data Cloud offers its customers the ability to enrich customer profiles by appending additional information to existing datasets. In other words, imagine using both lifetime value (LTV) metrics and social engagement metrics to truly understand who your most loyal customers are.
Segmentation
There is no point in having all this data if you’re unable to use it. Data Cloud offers users the ability to create Segments using all of the data ingested by Data Cloud. Segments and Calculated Insights — multi-dimensional metrics (i.e. calculate LTV by summing all completed orders) — can then be pushed into external systems such as Marketing Cloud for future use.
Better yet, Segments and Calculated Insights can be created without needing knowledge of SQL, although there are limitations as I’ll go on to explain.
Real-time Data Updates
If you’re in need of up-to-the-minute data and insights, Data Cloud might just be for you. Its streaming insights and real-time Data Streams allow users to work on the most up-to-date data instead of outdated insights and decisions.
This list is by no means an extensive list of Data Cloud’s capabilities but a list of what I believe are Data Cloud’s most useful tools. For a full list of capabilities, it’s worth checking out the product in more detail.
What are key considerations during Data Cloud implementation?
So at this point, Data Cloud sounds pretty fantastic. And don’t get me wrong, it is!
However, like any software out there, there are key considerations to take during Data Cloud implementation. Here they are.
Data Quality
Data Cloud is only as good as the data it is supplied. I was, and I’m sure among many, one of those Salesforce marketing enthusiasts who thought Data Cloud would solve all my data silo issues.
Data Cloud works on reconciliation rules, it uses these rules to unify data coming from different sources, so if your data sources don’t have commonalities between them, you’ll have a hard time creating your unified profile. Likewise, if your data sources are providing inconsistent data in each run, your Segments and Calculated Insights are only going to be so effective.
Set-up Complexity
Data Cloud is a very flexible platform, and it allows users to consume a variety of Data Streams and utilize a wide range of Data Model Objects. However, this also brings complications as it requires users to have a broad understanding of Data Mapping, APIs and Data Transformation, as well as having a solid understanding of Salesforce.
This is particularly important when Unified Profiles are involved. And furthermore, it requires a deep understanding of the platform’s very intricate nuances.
To list a few of the nuances:
- Unified Individual – The Unified Individual Object itself is non-editable. It essentially acts as a carbon copy of the Individual Object and is only created once Reconciliation Rules have been set up.
- Activations – Only fields that are mapped to the Individual Object are available as fields in segments pushed from Data Cloud. Related Objects and their fields are pushed as parsed fields, which adds complexity for using tools like Marketing Cloud Engagement.
- Profile Explorer – The OTTB Profile object is very limited and will require a lot of Salesforce expertise to build a usable page for viewing Unified Customers.
- Learning Curve – Data Cloud is a data-heavy tool and, at least in my experience, usually falls under the MarTech umbrella. Whilst this provides marketers with data-driven insights and segments, it also means a lot of learning is required.
As mentioned above, there are new concepts as well as new terminology such as Data Streams, Data Bundles and Data Lake Objects, but the biggest learning curve will come from the Segmentation and Calculated Insights.
Whilst Data Cloud does offer a ‘Builder,’ creating both insights and segments using the Unified Individual (the main reason for using Data Cloud) is achieved via SQL due to how the Unified Individual reconciles multiple profiles from multiple sources.
Use these tips for a successful Data Cloud implementation
I don’t want to sound like I’m being negative as, in reality, Data Cloud is a fantastic tool and can help drive meaningful engagement. But I do want to stress that understanding the detailed capabilities and key considerations of Data Cloud is the only true way of ensuring your Data Cloud implementation will be successful.
Questions to ask before Data Cloud implementation
If I was procuring Data Cloud for myself, I would consider the following;
- Where is my data coming from? If the majority already sits within a Salesforce product, then the chances are I can get a unified customer profile through smart architecture.
- What is the state of my data? As mentioned, Data Cloud won’t fix your data health issues. If your data is generally incomplete and lacks consistency, then you’re not going to be ready for Data Cloud yet — it doesn’t mean it won’t be right in the future.
- Who is going to own this product? It’s often marketers who will benefit from the segments and insights. But more often than not they don’t have SQL experience. If you’re hoping the segment and insights builders will make up for a lack of SQL knowledge, it might be worth reconsidering.
Need help with your Data Cloud implementation? Reach out to the Sercante team who can walk you through it and get you the results you’re trying to achieve.
