Customer data platforms, or CDPs, and data management platforms, or DMPs, are often confused with each other.
Customer data platforms, or CDPs, and data management platforms, or DMPs, are often confused with each other.
This confusion is understandable since they both use data to build audiences for marketers. What separates the two platforms is the kind of data each one uses. Those different types of data mean that the two platforms have to be used in different ways.
A customer data platform is best used to collect and organize data from the users on your website, mobile app, or anywhere else someone might interact with your business. A CDP can then take that data and share it with the other tools in your stack. One benefit is that this can improve your targeting, which makes your marketing more relevant to specific segments of your audience.
For example, you can take the data collected by your CDP and share it with your email marketing system to send relevant emails to one segment of your audience.
A data management platform, on the other hand, is a platform that collects and manages large, anonymized data sets of audiences. DMPs get their data by purchasing it from a data seller or by having such a large number of users that the DMP can aggregate and anonymize its own data; that is essentially how Facebook works. Facebook has billions of users; they collect data on their users, anonymize it, and sell it to advertisers.
The data DMPs collect is most often used to create advertising audiences. Facebook is one of the best-known examples of a DMP. Their DMP works by collecting data on their users and allowing advertisers to use that data to target Facebook users with advertisements.
Whether you should use a DMP or CDP depends on what kind of data you plan to use, how long you plan to retain it, and how you'll use customer identities as part of your business.
CDPs and DMPs both use data, but the types of data they use differ drastically. CDPs primarily use first-party data and a bit of second-party data. DMPs mostly use third-party data, also with a bit of second-party data.
These three types of data differ in how they're collected and shared:
This is data collected by your company. For example, someone visits your website and subscribes to your email newsletter. Their email address is a piece of data that was collected by your company. It is first-party data.
First-party data is generally regarded as the most valuable data and the safest type of data to collect because you can create a trail proving exactly where it came from and why it was collected. Proving that your data was collected ethically will lower your chances of privacy violations.
This is first-party data that is collected by another company and shared or sold to a noncompetitive partner. For example, if you decide to partner with another company to create an ebook, they might choose to share their email list with you so that you can distribute the ebook. Their email list is their first-party data, but to you it's second-party data.
Since this type of data comes from a partner, it's not as high quality or as safe as first-party data because you can't prove it was collected in an ethical way. Your partner can prove that, but you cannot.
This is data that is collected by a data-collection company and then shared with anyone who wants to purchase it. If you've ever run ads on Google Display Network, you've used third-party data. Google tracks the websites that someone visits and then buckets those people into broad advertising categories. If someone visits a lot of travel websites, Google will put that person into the “frequent travelers” bucket.
Third-party data is generally the worst quality data to use. Data-collection companies typically don't verify, or even guarantee, the accuracy of their data. On top of that, it's hard for you, as a marketer, to prove that this data was collected ethically.
First-party data tends to be the most useful and valuable for companies. Third-party data can still be useful, but the fact that third-party data is available to anyone makes it less valuable: You and your competitors could be using the exact same third-party data to run your marketing campaigns. That's not possible with first-party data.
There can be some overlap between second-party data for both CDPs and DMPs. It's very rare that a DMP will use first-party data, and it's very rare that a CDP will use third-party data.
For DMPs, that second-party data comes from a data exchange. Many DMPs are starting to build data exchanges where anyone can buy and sell company data.
For customer data platforms, that second-party overlap usually comes from one tool sharing data with another tool. For example, the tool you use to collect emails on your website collects those emails as first-party data. When those emails are shared from your email collection tool to your CRM, that data becomes second-party data to your CRM.
The biggest differentiator for customer data platforms and data management platforms is their use of customer identities, or personally identifiable information (PII).
Generally, PII is any piece of data that identifies a specific person. It could be a Social Security number, a phone number, an email address, etc. In marketing terms, PII is usually a combination of data points that can be used to identify one specific person.
The whole point of a CDP is to collect as much information as possible on every person in your audience. The more information your CDP collects, the more useful it will be, because it will allow you to make your marketing more relevant to your audience. For that reason, CDPs rely on PII to operate.
CDPs can process almost any type of personally identifiable information. It all depends on what you want your CDP to collect. It could be a full name, an email address, transaction data, social media interactions, and more.
The key to collecting this PII is to work with a CDP that ensures you're collecting PII in an ethical way. Good CDPs have very strong data security and will help you set data standards to make sure that website visitors have opted in to your data collection. Once visitors have opted in, your CDP also needs to give you the ability to delete visitor data if a user asks you to do so.
DMPs, on the other hand, use anonymized data. As a result, DMPs don't use PII, but they still collect data. That makes it harder to confirm that the data was collected in an ethical and legal way.
Another big differentiator between CDPs and DMPs is how long each of these platforms retains data.
Most CDPs give their users the ability to retain customer data for a long time. Smart CDPs even let the user set limits on how long their customer's data is held, and many CDPs allow their users' customers to request access to or deletion of their data.
CDPs retain data for longer periods because they become more useful when they have more data. The additional data allows CDP users to make their marketing more relevant to their customers.
For example, let's say you run marketing at an ecommerce company that sells furniture. At the end of the year, you decide you want to run a promotion to give a 15% discount to your most valuable customers from the past year. How do you figure out who your most valuable customers are? You can use transaction data stored by your CDP to find the customers who spent the most money with your company in the past year.
DMPs are the complete opposite. They're best when they only retain data for a short time — 90 days is common. Think about it: If you're an advertiser looking to target people interested in travel, you don't want to target someone who was interested in travel five years ago. You want to target people who are interested in travel right now or have been interested in the past 90 days, at most.
Let's look at a few use cases to see how customer data platforms and data management platforms are used. We'll base these examples on a fake SaaS company called Zwix, which develops project management software. Zwix uses both a CDP and a DMP.
Since Zwix's CDP monitors all website activity, it captures data when a potential customer visits Zwix's website from LinkedIn. The CDP tags that person as a LinkedIn visitor, even though there isn't any other PII yet. That potential customer then reads three blog posts. The CDP captures that data, too. The potential customer leaves Zwix's website without taking further action.
A few days later, that potential customer comes back to Zwix's blog. Zwix's CDP has the ability to recommend other blog content based on what that potential customer previously read. The potential customer reads a few more blog posts and decides to subscribe to Zwix's newsletter.
Zwix's CDP captures this email and ties it to the original referral source and all website viewing behavior. This step, called Identity Resolution, ties together all interactions the potential customer has taken with your website and creates a user profile.
Now, Zwix's CDP can add that potential customer to an email audience based on specific blog posts viewed and can then use relevant messages to push that person further down the marketing funnel. Zwix can also add that potential customer to an advertising audience on Facebook or Google Ads and retarget that person on those platforms.
Since DMPs are best used in digital advertising, Zwix uses DMPs from companies like Facebook, LinkedIn, and Google to build audiences that they want to reach on those platforms. Those platforms collect data from their users, categorize it, and make it available to advertisers.
Zwix uses LinkedIn's DMP to build an audience of project managers who work at Fortune 500 companies by adding that criteria to their target audience. LinkedIn then serves ads to that target audience.
Zwix also uses Facebook's DMP to build a look-alike audience. Facebook allows advertisers to upload an email list of their customers. Facebook takes that list and finds other Facebook users who are similar to the customers on the list. Facebook then serves ads to Facebook users who are similar to Zwix's customers.
Both CDPs and DMPs can be very useful platforms. Deciding which one to use depends on your goals, but there is some overlap between the two platforms.Use a DMP if you want to build marketing campaigns for audiences that are unfamiliar with you. DMPs are best for this because of their use of third-party data. They can give you access to audiences that you don't know. You can then use that new data to build a targeted marketing campaign.
CDPs are built for gathering, sorting, and dispensing first-party data. CDPs can also be great for ethically collecting that first-party data, and proving that you followed proper protocols to collect your data. If you plan to create highly personalized marketing campaigns based on your own data, use a CDP. Your CDP can gather website data and send it to any number of different tools, depending on your needs.
Ultimately, there's a chance you're going to use both platforms to some degree in your marketing. Your CDP can handle the heavy lifting of monitoring and categorizing website visitors, while you might use a DMP to take some of that data and use it to build advertising audiences.
No matter what platform you use, make sure the data you're handling has been collected in a legal, ethical way.
If you're ready to take the next step and find a CDP that's right for you, download our CDP Buyer's Guide.