I had the opportunity earlier today to sit on a panel at RampUp from LiveRamp specifically dedicated to the topic of customer data platform’s status as the newest shiny object. Conor McKenna, VP at investment banking firm Luma Partners, teed up the panel discussion with an overview of the space and how Luma sees it. There were immediately questions, both for Conor and then for the panel, that I thought I’d document and answer more thoroughly than we had time for at the session. So, here’s my take:
1. How do CDPs handle transactional data?
On its surface, this seems like a really simple question – it’s customer data so of course, a CDP should handle it. But it’s actually more complex than many marketers realize. One reason is that historical transactions and offline transactions are typically stored in ERP-type systems. This means that different CDPs will actually handle transactional data quite differently. Typically, they are handled in three ways.
- Some CDPs will incorporate transactions into an events stream and map them to an “identity graph” or similar entity, but segmentation can be problematic.
- Others, with more relational, rigid SQL databases will tie transactions to customer records but will struggle to connect that data with unstructured sources like web data.
- Lastly, some CDPs – and BlueConic is in this camp – convert this data to customer attributes by flattening the transactions and data associated with the transactions (e.g. product category, purchase date).
Which option is best for you will depend on how much data you have, what you’re trying to do with it, and where the data is coming from.
2. What about ROI?
How about ROI indeed! Again, this can be really simple for the one trick pony CDPs who are expressly solving for one data/marketing challenge. For example, one CDP promises an increase in cart conversions in the first four weeks because that’s the use case they go to market with for everyone. Another CDP, rebranded as such after 15 years in campaign management, has 9-12 month deployments so ROI calculation is going to look quite different. Implementation aside, the impact of a CDP can broadly be bucketed as either efficiency or monetization; but it’s marketers that need to plug in the numbers to the equation in order to make sure they’re seeing an ROI. What’s the price of saving your data specialist one full day per week because you don’t need to request a new list every time you make a segment change? How much is a new identifiable visitor whose email address you were able to capture via personalization on your website worth?
3. Do CDPs and CRMs bleed together?
In the sense that both deal in customer data, absolutely. And should! For example, our own marketing team at BlueConic has dozens of attributes mapped between BlueConic and Salesforce, our CRM and system of record. Since the two systems sync instantaneously, we can interact in real-time, whether that’s on the website or via a sales rep following up with prospective customers directly. This piece in CMSWire explains the relationship well.
4. How much are CDPs starting to integrate with ad players?
“Ad players” is, of course, a very broad term. The short answer is that CDPs are integrating and have been for some time with a variety of ad players. Whether that’s DMPs, or directly with AdWords or Facebook, or with platforms like, you guessed it, LiveRamp. While the technical capability is there, I do think the adoption lags. But the results speak for themselves; ask Franklin Sports who has seen a 360% return on ad spend using BlueConic to refine their retargeting campaigns.
5. Can you speak more specifically about the differences between the CDP and DMP?
We can and have, as have others so let me defer:
…and a few more:
- “What is this thing we call a CDP?” by Marty Kihn, of Gartner at the time, for AdExchanger
- “Move over DMP, DSP, and SSP: CDP Is AdTech’s Hot New Acronym” in Digiday
- A discussion on Quora about this very topic
6. How should marketers be thinking about whether or not a CDP is right for them?
All marketers want to better utilize first-party data and we believe CDPs are the means to that end (you’re shocked, I know). But achieving this goal depends on your business, priorities, marketing goals, resources, and existing marketing tech stack. Not every CDP is going to be right for you. For example: if you’re trying to map every single data point – deterministically, probabilistically, and algorithmically – is your priority in search of a golden record, then you want a solution like Reltio that is more akin to master data management. On the other hand, if personalization is your top use case and the data is secondary, then maybe you want to look at an Evergage, which has some CDP light features next to their magic quadrant-leading personalization engine. Solving for the ways in which first-party data can make your marketing better and more efficient should be on every marketer’s priority list – but which solutions enable that will vary widely.