Personalization Engines versus CDP
|Disclosure: The links on this page are "Affiliate Links" and while these are shown at no costs to our viewers, they generate commissions for our website(s)|
Personalization Engines versus Customer Data Platforms (CDP) – What’s Inside?
1. Customer Data Platforms – What is a Customer Data Platform?
2. What Does a Customer Data Platform Do?
3. What Kind of Data Do CDPs Collect?
4. How Does A CDP Work?
5. Who Is a CDP For?
6. Types of Customer Data Platforms
7. What a CDP is not?
8. Personalization Engines
9. How Do Personalization Engines Work?
10. Examples of Personalization Engines Use
11. What’s the Difference Between a CDP and Personalization Engine?
13. What Personalization Engines do that CDPs Don’t
14. Which is Best for You?
Since the dawn of marketing, marketers have been on an endless journey to find the best ways to capture relevant customer data to help them deliver the best customer experience based on what customers want. Today’s businesses have a massive marketing advantage compared to marketing toolkits of the not so distant past.
Now, more than ever before businesses can get an intimate view of the customer journey. Not only can they see in real-time what customers are doing, but with the help of Artificial Intelligence and Machine Learning marketers can predict what customers want and deliver it when they want it and the way they want it.
This article is going to focus on two technological innovations that have created this new marketing landscape – Customer Data Platforms (CDP) and Personalization Engines . To compete in the digital age businesses need to be aware of what tools have changed the industry – what they are, what they do, how they’re different, and where they’re taking businesses and their customers.
Let’s dive in.
It’s always best to start with definitions. A Customer Data Platform (CDP) is a software that aggregates and organizes customer data across a variety of touchpoints and is used by other software and systems for marketing efforts.
This innovation frees marketers from having to depend on a technical specialist to pull data for them when they need it. The software collects the data, organizes it, and makes it accessible.
The CDP Institute defines CDPs as, “A packaged software that creates a persistent, unified customer database that is accessible to other systems.”
Customer Data Platforms are like a three-in-one power tool for the marketer’s toolkit. According to The CDP Institute, a CDP comprises three elements:
The CDP is a prebuilt customized system configured for the needs of the business or marketer.
Persistent, Unified Customer Data Base
The CDP creates comprehensive customer profiles by capturing data from multiple systems and linking information related to the same customer. It then stores the information and tracks behavior over time.
Accessible to Other Systems
Data stored in a CDP can be used by other systems to analyze and use for targeted marketing purposes.
Most captured customer data originates in separate systems that are not designed to share. They tend to operate in silos. Methods like using enterprise data warehouses to collect data into unified customer profiles have failed to solve the problem. Approaches like “data lakes,” have been able to collect data, but have failed to effectively organize it for easy access.
A Customer Data Platform has solved the sharing and organization bottlenecks of prior technologies. CDPs give marketers direct control by providing them a holistic view of the collected data. CDPs aggregate and organize personas into the most likely or relevant audience for a specific product or service. They help businesses find the most likely customer to target for their product or service.
CDPs are able to collect customer data that is left behind as customers interact with companies online and offline through websites, blogs, e-commerce portals, and in-store interactions. There are four kinds of customer data CDPs collect and organize for marketers.
This data builds the foundation for every customer profile. It helps businesses to uniquely identify each customer and avoid creating replications. Identity Data includes information about:
- Phone and email
- Social Media
- Job Title and Employer
- Account info – Like an Amazon ID and account number
Expanding on identity data this data gives businesses a fuller picture of their customer. Descriptive data can vary based on the type of company. For example, a car dealership may collect details about a customers’ car choices, while a baby formula company would collect data about the number of children in a customer’s family. Descriptive data can include information about:
Behavioral or Quantitative Data
Behavioral data captures how customers engage with a business through specific actions, reactions, and transactions. Behavior data includes information about:
- Transactions – Number and type purchased or returned, abandoned carts, and order dates. This metric also captures the timing of purchases, frequency, and amount spent.
- Email – Email opens, click-throughs, responses, and dates.
- Online activity – Websites visited, click-throughs, product views, and social media engagement.
- Customer service – Communication dates, query details, and service rep details.
Qualitative data provides context for customer profiles and personalizes it by capturing motivations, opinions, or attitudes.
- Motivations – This is information collected from surveys with questions like, “How did you hear about us?” and “Why did you purchase this product?”
- Opinions – Again, this might be collected from surveys with questions like, “How would you rate this product?” or “How would you rate our customer service?”
- Attitudes – Things like favorite color, food, or political party preference informs this data point.
Now that you know the type of data CDPs can collect you probably want to know how they work, or how they leverage that data for marketing purposes.
CDPs focus on marketing to a known audience. The goal is to increase conversions, retention, and engagement with an existing customer base. CDPs primarily use PII (personally identifiable information) and first-party data, although third party-data can enrich a CDP. Personally Identifiable Information, or PII, is any data that could be potentially used to identify an individual. Examples of PII include full name, social security number, driver’s license number, bank account number, passport number, or email address.
Companies use CDPs to keep user data in one place and then access that information to implement personalized marketing strategies across multiple channels like the web, ads, email, and mobile devices.
The work, or heavy lifting, that CDPs do for companies comes in the form of organizing and de-siloing data into a centralized and accessible space. This saves companies countless hours of integration work they once had to dedicate resources to. Not only that, CDPs allow various tools and other systems within a company to easily integrate the captured and organized data for specific marketing channels and campaigns.
CDPs are the ultimate tool for collecting, organizing and sharing critical customer data.
CDPs are used by marketers and B2C businesses for the most part. A Forbes survey reported that 78% of organizations either have or are developing a customer data platform . Most of the businesses using CDPs are larger enterprises that manage hundreds or even thousands of employees.
There are many types of CDPs, because they are customized to a great extent to the organization’s needs. There are four categories you should be thinking about when determining what you want a CDP to do.
CDP vendors differ in their privacy, security, and governance standards. Many CDPs for mid-market organizations will not offer partitioned storage, role-based permissioning, or controls for EU privacy compliance – which are all critical for working on a global scale.
CDPs vendors offer different types of integrations. You need to be sure the CDP you’re buying supports platforms and tools that matter to your organization.
A number of CDPs are not well-suited for mobile and omnichannel marketers. If your customer journey includes people interacting via mobile platforms or connected devices you’ll need a CDP that supports these.
Remember, CDPs organize and distribute “first-party” customer data. However, if you wish to learn more about your customers from other sources, your CDP will need to be capable of facilitating data from partners and third-parties.
CDPs can do a lot of things, but they can’t do everything that current marketing technologies offer. CDPs alone, are not able to do predictive and decision modeling and some don’t take action on customer insights.
New add ons are emerging in the market to help with this. But currently, alone, CDPs don’t offer the machine learning capabilities that Personalization Engines offer. We’ll get into that shortly.
Additionally, CDPs are not:
- Customer Relationship Management Systems (CRMs)
- Data Management Platforms (DMPs)
- Digital Personalization Engines (DXPs)
- Master Data Management Systems (MDMs)
However, many of these systems can work well with a CDP.
Customers today are demanding brand interactions tailored to their individual needs. Today’s companies are operating in an Uber-like generation where greeting people by name in emails or injecting subject-line personalization are table stakes. Consumers today expect a truly individualized experience based on their intent and needs, where and when they want it.
This change in attitude has driven the rapid growth of personalization engines.
The main idea behind the personalization process is to create a relevant, individualized interaction between consumers and companies to enhance the customer experience. This is achieved by using insights from individual behavioral data combined with data of similar individuals to deliver an experience that meets specific needs and preferences.
Marketers use personalization engines to ensure customers see the information they want to see, in a way they want, and when they want. Now, personalization has been around for a long time and what it means to a business in terms of customer engagement and experience has evolved over a period of time. Basically, it is no longer enough to just personalize emails and greet people by their name; best-in-class companies will need to compete on experiences, not just on products and services.
For example, if a college student visits a website, she may see more video options, or a person who is browsing the mobile app of a restaurant in an area where the weather is cold will get information about hot beverages. Likewise, a visitor who likes the San Francisco 49ers may see promotions for merchandise for their favorite team while another user on the same page may get served products for the Kansas City Chiefs.
To achieve this level of relevance at scale, companies are challenged to think beyond segmentation, batch analytics, A/B testing or rules-based interactions. Such a level of personalization requires the ability to collect and make sense of the ever-changing data crumbs that customers leave, make smart interpretations, and deliver meaningful interactions so every visitor and customer feels valued.
Personalization engines are all about predictions. That’s the magic sauce that makes them effective. But how do they predict?
Personalization engines collect data, then use the power of Machine Learning to make predictions about what the customer wants, intends, purchase propensity, and so on. Machine learning is a field of computer science that is roughly defined as computers “learning” from past experiences and observations.
Predictive programming combines data, code, and science to find patterns that rule-based programming cannot find. Instead of creating a program based on rules, an algorithm is programmed to find opportunities based on mining data to produce decisions.
The key to making this happen is identifying consumers across channels, devices, and visits. While CDPs also provide companies a complete customer profile, advanced personalization engines go a step further by providing the customer’s dynamic profile. This means that they take into account:
- The consumer’s attributes and demographics
- The consumer’s past behavior
- The consumer’s current activity either in the store, website or mobile, or any other device. This capability helps in personalizing the experience of unknown users as well.
- The consumer’s current context includes their location, weather, special events, and so on. This ensures that the experience every visitor receives is truly personalized to their needs.
By optimizing for each visitor, instead of segments, the machine is able to deliver at a more individualized level than what marketers are able to accomplish manually. The more data that is accessible and analyzed, the more accurate the decisions become.
However, this does not mean that the first step for a company is to create data lakes or collect and store every bit of data they can get on their customers. Advanced personalization engines have the ability to access the right kind of data that will help the models to make quick and right decisions about the kind of engagement required for every visitor.
Some retailers use personalization engines to personalize the order of its navigation menu based on user’s preferences. Because navigation menus serve as roadmaps for product discovery, serving customers with optimized and reorganized menus decreases the time needed for users to find what they’re looking for and more effectively drives conversions.
Another effective way retailers use personalization engines is offering urgency messaging to encourage coupon redemption or to make a purchase. A customer may be on a website looking at a product and in real-time they will be presented with a time-sensitive coupon for the very item they’re considering. The promotional discount will be automatically added to their cart encouraging them to complete the purchase. Others might get a notification about their reward points as they enter a store, or the in-store availability of the item that is in their digital cart.
Delivering such highly targeted and personalized experiences is powerful as it is based on the analysis of that particular visitor’s unique interests and activity. It allows companies to maximize customer engagement opportunities with their brand.
Now you have a good idea about what CDPs and Personalization Engines are, how they work, and how they benefit marketers. While there is some overlap between the two, the main thing that Personalization Engines do that CDPs don’t do is apply the holistic context of individual users and then deliver a personalized experience or offer based on predictive modeling. Personalization engines are personal, as the name implies, but even more, they are decision-makers. Personalization Engines leverage machine learning to automate and deliver an extremely personalized customer experience that no other system or human can come close to achieving.
Personalization engines and customer data platforms are both powerful tools that have transformed the marketing and customer experience. However, it is good to remember that CDPs alone aren’t the solution to hyper-personalization, but they do make the process of aggregating and analyzing data much easier. Most industry leaders agree that true personalization that fulfills the expectations of today’s customers requires AI and ML that only advanced personalization engines possess.
Content provided by Simona Rich