Composable CDP

What is a composable CDP?

A composable CDP (customer data platform) is a marketing technology infrastructure that sits on top of a cloud data warehouse and activates customer data to various marketing and sales channels.

Since composable CDPs are built on the foundation of a company’s cloud data warehouse, they pull information from a single source of truth for all customer data. The composable CDP gives marketing teams a self-serve (low- or no-code) interface to create and segment audiences on top of the data warehouse. Then, it allows marketers to activate those audiences in real time to channel destinations, such as CRMs, email marketing tools, and ad platforms.

Traditional CDP vs Composable CDP

A traditional CDP is an all-in-one software solution that includes all the components to collect, classify, store, and activate customer data. That means a traditional CDP, also called a packaged CDP, stores a copy of the company’s data and dictates which audience channels can sync with that data. 

By contrast, a composable CDP is an activation layer that fits on top of your company’s existing tech stack, rather than operating as a separate entity. Essentially, it integrates with your marketing tools and helps you unlock new customer insights from your existing data — in other words, take action on your data to drive more personalized customer interactions.

‍

‍

Where did composable CDPs come from?

Marketing cloud

Cloud storage started gaining popularity in the early 2000s as a way to securely store large amounts of data. The marketing cloud followed soon therafter as the original all-in-one solution that managed customer data, journey orchestration, segmentation, and business intelligence. 

Nevertheless, the marketing cloud solutions only helped compile data from its own marketing platforms, which meant much of the customer data was still stored across platforms. The marketing cloud couldn’t incorporate data from a customer’s activity with a product, for example. This siloed approach made it difficult for marketing teams to use that data for meaningful campaigns at scale. 

Traditional CDPs

The concept of a customer data platform originated in 2013 as an all-in-one solution to collect and activate customer data across various touch points — marketing, sales, customer success, product, etc. This offered the ability to build a 360-degree view of each customer and tailor marketing activities to that customer’s interactions across different touchpoints. 

However, traditional CDPs pose several issues for marketing teams:

• Engineering teams must set up and maintain the code for accessing data, and marketers can’t easily export the data on their own.
• Data only syncs with end platforms reactively and is not stored centrally for proactive audience building.
• Data is only collected when the system can identify the user with a cookie after they took an action on a website or mobile app.
• Data cannot be collected from offline activity.

Composable CDPs

A composable CDP is built on top of your data warehouse. This lets you directly activate the data you have, where you have it, to generate the insights you need. Then, you can build audiences with that data and sync them with your existing martech stack, such as CRMs, search ads, social channels, and more.

‍

‍

How do composable CDPs work?

Composable CDP features

A composable CDP has the same core features as a traditional CDP, the difference being most of those features are your existing tech stack. These features include:

• Data ingestion - This is the first component, where customer data across multiple sources is collated into a unified customer profile. These sources could include event-based data (such as email signups) or first-party data attributes (like income level or age). Snowplow and FiveTran both offer comprehensive data collection functionality.
• Data storage and modeling - This component involves aggregating the customer data into a data warehouse and then defining your customer audiences. You can use custom AI models to create these audiences, or lean on pre-built models provided by the composable CDP.
• Data activation - The final component involves sharing those customer audiences and other relevant data points with your marketing channels. Composable CDPs specialize in this step, also known as data activation.

‍

‍

‍

Composable CDP architecture vs traditional CDP architecture 

In a traditional customer data platform, all the above components come together in a single package. With a composable CDP, because it sits on top of the existing data warehouse and integrates with existing marketing tools, teams can choose among the best vendors for each component, depending on their goals, budget, and existing tech stack. 

Why should I care about composable CDPs?

You may be wondering why you should care about composable CDPs and how these platforms could help your team. Below are a few key benefits of composable CDPs.

See complete picture of the customer

All CDPs integrate customer data from multiple sources. However, a composable CDP integrates data from all datasets currently in your warehouse — not just sales or marketing information - including demographic attributes like age or gender, as well as lead score, last product login date, lifetime activity, and so on. Therefore, you get a more comprehensive picture of the customer than with traditional CDPs, which mostly rely on behavioral datasets like email signups or page visits.

Use best-in-breed marketing platforms

A composable CDP lets you use the most appropriate platforms for data collection, storage, and activation depending on your specific marketing goals, rather than being tied to what a traditional all-in-one CDP or marketing cloud offers. 

Avoid disrupting your marketing stack

With a composable CDP, teams don’t have to continuously build and maintain separate data pipelines. It integrates with your existing martech stack, making it easier (and less expensive) for all teams to access customer insights without any extra steps.

The advantage of a composable CDP is that teams can add and modify elements as their needs evolve and their audience knowledge grows. For example, you may start with only email marketing and social media channels but eventually add push notifications and search ads. 

They’re future-proof

As your marketing strategies evolve and more advanced tech hits the market, you can add pieces on top of your existing composable CDP structure rather than setting up a whole new stack from scratch.

How do marketers use composable CDPs?

More nuanced audiences

A composable CDP activates all customer data in your data warehouse. The more data you have, the more granular you can get with audiences, allowing you to create more personalized campaigns. For example, you can more easily create a specific winback audience of customers whose last login was more than twelve weeks ago. You could also develop lookalike audiences based on your best customers, or share “related product” recommendations with customers who’ve just browsed or bought certain products.

Meaningful customer journeys

In some cases, the composable CDP can help you to set up custom workflows based on the audiences you define. For instance, you can auto-enroll customers into different email marketing campaigns depending on product activity. A user that only focuses on one area of your product may benefit from a sequence that showcases another feature they may not know how to use. 

You can also create audiences based on buying history and share upsell or cross-sell campaigns with customer categories who are most likely to make another purchase. 

It’s important to note that not all composable CDPs offer this functionality.

Testing and experimentation

Using thesingle source of truth for customer data helps you measure the actual results of your campaigns: from revenue to app downloads and any other metric in your data warehouse. Be able to tweak customer experiences and campaigns based on granular, real-time insights, not just performance metrics in marketing tools. With real-time data, your team can create new tests and iterations to optimize campaigns.  

Will a composable CDP affect my customer data privacy and security? 

With a composable solution, you retain complete control over your customer data. The composable CDP sits on top of the data warehouse and queries in place, so there are no data copies. Only the properties you need are sent to the activation platforms. You have control to define what data is shared at every stage. 

And since you’re using your own data warehouse, you get to apply all your existing security and data governance protocols. This is a major benefit in a world where data legislation is constantly tightening and consumers are increasingly conscious about where their data is going. 

How can my team implement a composable CDP? 

To get started with a composable CDP, you’ll need data ingestion, collection and storage in a data coud. From there, you’ll need to activate that data by building audiences and syncing with various destinations (email, social media, ads, etc.). 

Data storage

A composable CDP sits on top of the data cloud - the central location for customer data. You can store this data in your data warehouse, such as BigQuery or Snowflake. You can also use a data lake, such as AWS Athena or Azure Synapse, as your data storage source. 

Data collection and ingestion 

Once you know where your customer data will live, you’ll need to find a way to get data from your first-party data sources into your data warehouse or storage platform. For example, you may need to take data from Salesforce and put it into a data warehouse like BigQuery. A data ingestion tool, like Fivetran or Stitch, can help you do this.

Data organization

After your raw customer data gets into the data warehouse, you’ll want to organize it to make it useful for your team. This includes:

• Identity resolution - Identifying the same users across different data sources. 
• Data models- Create a final and clean view of your customers and the various data points you have for the customers. 

You define which datasets you wish to connect to your composable CDP. You can then set permissions based on the outcomes you’re driving (e.g. only customers in certain demographic bands), or exclude datasets that aren’t relevant to your current marketing goals. 

Building audiences and user journeys

Once the data is organized, composable CDPs help marketers to create audiences and map user journeys across various channels and destinations. 

Since a composable CDP sits on your data warehouse, there’s more flexibility with the customer data points you can use and track. 

Destination sync

Once you have audiences built and journeys mapped, a composable CDP allows you to activate those audiences to different destinations — CRMs, search ads, push notifications, etc. Traditional CDPs typically prescribe specific audience destinations, while composable CDPs can use the platforms your team already has in place.

Choosing the right composable CDP vendor

There are a few important factors to consider when selecting a composable CDP vendor. 

Compatibility with your industry 

The first consideration when choosing a composable CDP vendor is how much experience they have working with your business type or industry. For example, marketing needs vary greatly between the retail and healthcare industries — especially when it comes to data privacy and security. Similarly, large, enterprise-level organizations have very different needs than small startups.

Compatibility with your use case 

Before settling on a composable CDP vendor, your team should define their use cases for the platform. Use cases can include: 

• Creating hyper-personalized customer journeys to up-sell and cross-sell products 
• Developing effective and informative onboarding sequences
• Crafting highly targeted marketing campaigns for prospects
• Optimizing the omnichannel experience for customers and prospects
• Deploying re-engagement or win-back campaigns for past customers

Once you know your specific use cases, verify that the composable CDP vendor has experience in those areas. Reading the user reviews or talking to current/former clients is a great way to learn about the platform’s capabilities. 

Integrations offered

The composable CDP vendor should offer all the integrations needed to accomplish your target use cases. These integrations could include: 

• Advertising tools
• Analytics tools
• Business intelligence tools
• Customer service platforms

Moreover, the vendor should be continuously adding new integrations to keep up with evolving needs. 

Security features 

You’ll want to verify that the composable CDP vendor has an industry-standard, independent security certification like SOC 2 or ISO 27001. Data security is paramount, especially when dealing with sensitive healthcare or financial data. Certifications like these demonstrate that the composable CDP vendor is continuously evaluating and upgrading its security practices.

Onboarding and customer service 

One of the major benefits of a composable approach is that it doesn't require as much internal engineering time as a CDP, but you will want to confirm that when looking at vendors. You should be able to fit the CDP solution seamlessly into your tech stack and start activating data right away. The composable CDP vendor should offer support throughout this process, as well as ongoing assistance for any product upgrades or support with new use cases you might encounter as your organization evolves. 

Composable CDP examples

A sports organization stores its fan and audience data in a cloud data warehouse like BigQuery. With a composable CDP, the organization's marketing and sales teams can use a no- or low-code audience builder to create segments based on data in BigQuery. Then, the marketing and sales teams can deploy these segments across campaigns — testing different channels and creative as needed.

For example, the team could use information in the data warehouse to identify the fans that are likely to purchase season tickets (such as fans who have purchased more than 10 tickets in the last 90 days). Then, they could sync those audience segments with a tool like Salesforce to create opportunities for pitching a season ticket package.

‍