These past few weeks have given Customer Data Platforms a lot of press in the marketing technology world following the announcements from the three big marketing clouds that they will be adding CDP-like components to their “integrated” cloud solutions.
The first announcement came on March 25th with the U-turn from Salesforce, having previously debunked the CDP as "a passing FAD" in 2018. Although it's great that Salesforce now recognizes the importance of unifying customer data, the article focuses more on the existing activation and execution tools that it has acquired over the years.
So what does this mean for the Customer Data Platform market?
To be honest, only time will tell, and the market is waiting for future announcements from the big three to provide more clarity on what they will be bringing to the market.
However, the suspicious ones (and rightly so) point out that it is very likely that these CDP offerings will not fully meet the definition of a ‘RealCDP’ as outlined by David Raab and the CDP Institute last week.
Raab points out that there are five key criteria that a CDP must meet to be considered a ‘RealCDP’:
- Ingest data from any source
- Capture full detail of ingested data
- Store ingested data indefinitely (subject to privacy constraints)
- Create unified profiles of identified individuals
- Share data with any system that needs it
So let’s take a look at the criteria in closer detail:
1. CDP must ingest data from any source
Firstly, this has been an issue for every marketer since time began. Traditionally the Marketing Clouds have partnered with database companies and Marketing Solution Providers with expertise in building and re-building marketing databases to satisfy the need for their clients to create a Single Customer View. For marketers to personalize and create multi-channel offerings, they need to be able to bring data into their execution systems that is consistent, holistic and trustworthy, and without the data piece being solved, many brands have struggled to realize the extensive cloud investments they’ve made. Up to now, a common use case for fully invested cloud customers has been to utilize a CDP, like BlueVenn UNIFY, to take data from the CRM, channels and other operational systems to then perform the unification of the data to push the Single Customer View back to the channels. Essentially, it satisfies the need for unified data without having to invest 4X the $$$ on a database agency.
Will an Adobe CDP easily ingest data from a Salesforce CRM to then push to Salesforce Marketing Cloud? What if you’re utilizing Adobe’s Magento eCommerce? Will Salesforce take rival technologies’ data into its own CDP?
These are questions that will become apparent over time, but stresses the point that a RealCDP should ingest data from ANY source and share data with ANY system that needs it.
2. Capture full detail of ingested data
It’s crucial for a CDP to bring in ALL the data! This can make connectors, feeds and some data sources particularly tricky, especially when highly customized by an individual brand. All the data points can be essential for the additional ‘activation’ use cases that marketers demand from a CDP, and therefore is another requirement that the CDP Institute defines for its qualification as a 'RealCDP'.
This need for all the data can often require a level of customization and ‘Services’, and for all the promise of CDPs there will always be a need for some services to ensure that all the data comes in. For the lean, mean software (not services) marketing clouds is this something they want to add to their repertoire? Balancing the software vs services model for a SaaS business requires a delicate management, and will the clouds want to put these costs and margins into their P&L? After all, isn’t that why the larger SaaS vendors have always neglected the data in the first place?!
3. Store ingested data indefinitely (subject to privacy constraints)
Another use case for CDPs is the ability to use the resulting Single view of the customer for analysis and intelligence. Essentially, to link every click, open, like, chat, transaction or visit, including all the expanded data relating to those data points (such as the source, time/date, tag, SKU, category etc), to an individual record – a memory of each customer if you will. That’s a lot of data when you’ve got many millions of customers and potentially billions of events and transactions! It’s imperative that the data isn’t summarized or removed when undertaking analysis or analytics, and this also will need to be addressed in the Cloud CDP offerings.
CDP’s are all too familiar with the privacy constraints born out of GDPR and other upcoming international legislation, and have already modified their offerings accordingly. One-click options to strip PII out of records that request erasure, ensures that you are still able to analyze what your customers are doing without breaking modern data privacy rules.
Similarly, most data-savvy marketing professionals understand the limitations of cloud-based digital execution platforms for handling offline data and the need for multi-channel brands to de-duplicate on potentially many variables. Questions about where the data layer and UI will sit within the Cloud architecture is something that will be interesting to see.
4. Create unified profiles of identified individuals
This is the secret sauce of CDPs. The core use case for a CDP is to unify data, and although all 'RealCDP' certified vendors provide data unification and matching of identities to create the unified profiles, they all differ. Some provide unification but no data cleansing, others provide enhancements and offline/online matching, but again others do not.
The matching rules, cleansing, enhancements, data quality and hygiene aspects of the Cloud CDP offerings will need to be scrutinized in depth as they will undoubtedly favor the digital-first nature of their existing execution platforms.
5. Share data with any system that needs it
Criteria number 5 is perhaps the most intriguing element of a 'RealCDP' that will define the Cloud CDP offerings. The very nature of a CDP is that it can make the optimized data and profiles usable within existing marketing, CRM and customer-facing systems.
Will Salesforce want their CDP to feed the Adobe stack? And vice versa?
CDP’s provide an open architecture to provide data back to channels and customer-facing systems that need it and therefore is a critical component – particularly for brands that are wedded to certain technologies, and prefer not to replace their entire marketing stack.
CDP Conclusions (or Confusions)
As I mentioned before in this article, only time will tell, before we know how closely the Cloud CDPs will align with the 'RealCDP' criteria definition. What we do know however is that David Raab, the CDP Institute and the RealCDP certified vendors have created a very intriguing market that has overtaken all expectations. Finally, customer data management has a true seat at the table having fallen by the wayside in recent times to ‘shiny new tech’. The result is that “Customer Data Platform” requirements have been appearing in almost every MarTech RFI and tender over the past few years, which has led the big three to make their CDP move.
Want to know more? Download the ‘Marketer’s Guide to Customer Data Platforms’ eBook
This blog article is an extract from BlueVenn’s ‘A Marketer’s Guide to Customer Data Platforms’ eBook.
Download a copy of the full eBook for more CDP advice, how a CDP differs from other data management technologies, or the different types of CDPs and what to look out for when researching technology vendors.