A customer data platform (CDP) is a vital software system that centralizes a wide variety of data from a company’s user base in a consistent way. A CDP replaces the need for dozens of different direct integrations among a company’s disparate tools with one central hub that acts as a single source of truth. CDPs are built to combine customer data from a wide range of touchpoints into rich user profiles. These user profiles can then be used for targeted lifecycle marketing, detailed product / behavioral analytics, or any other business function which relies on accurate customer data. For all these reasons, an effective customer data strategy is one of the drivers of high data trust. And it reveals the customer data platform as one of the cornerstones of any data-centric company’s tech stack.
Every CDP is composed of three main components:
Every CDP addresses the challenges of data capture, storage & transformation, and delivery in a different way. And if you haven’t already guessed, it’s possible to build your own CDP out of a collection of tools that perform these three primary functions together. While this parallel discussion on Packaged vs Composable CDPs is beyond the scope of the current article, please check here for more details.
And yet CDPs like those offered by Segment, Rudderstack, mParticle, and more go far beyond storing and organizing customer data. In this article we’ll share how an appropriate implementation of the right customer data platform can accelerate growth and streamline operations across every department in your company.
Table of Contents
The most important factor of any customer data platform strategy has nothing to do with technology, nor will any CDP list it as one of their unique benefits. But you’ll see it if you open an org chart of your team. Cross-functional collaboration is the most important factor in any CDP implementation.
Think of it this way:
Your CDP is different, precisely because it connects to every other tool in your tech stack. Small companies with less than 500 employees still average more than 150 different tools, and for enterprise teams that number quadruples. The right CDP implementation can help your team manage customer data from all these different tools now and into the future. This includes everything on your roadmap today, plus obstacles you haven’t even thought of yet. But this is only possible if everyone on your team has a seat at the table.
There are three main reasons why cross-functional collaboration must define any CDP implementation, which together describe every effective CDP strategy:
We’ve worked with hundreds of different companies on projects related to data governance, lifecycle marketing, product / behavioral analytics, and everything in between. In almost all these cases, it made sense to implement a CDP at the same time to ensure consistent customer data collection. But if key stakeholders cling to old reports, they’ll continue to have the same issues with data trust. A CDP is a great way to get everyone on the same page, as long as your team embraces that vision.
For many companies, their perspective shifts once they stop thinking about native integrations among all the different tools in their tech stack, and start asking whether those tools will work with a CDP.
Your data team will appreciate a CDP for its ability to aggregate different types of data for accurate analytics. In the same way, your marketing team will value a CDP for the visibility it will give them into the customer journey. The catch? These dream scenarios are only possible if you have complete buy-in from the team members responsible for setting up the CDP. And this goes beyond creating consistent event names. The people who set up your CDP - usually your data engineers - must understand how the CDP works, which teams will rely on this centralized data, and how they’ll use it.
If your engineers implement the customer data platform in isolation, how can you guarantee other teams will be able to self-serve that data to complete their work? Far too often, we see customer data siloed not only within different departments, but also across workflows. If your engineering team sets up the CDP without any input from your marketing or product teams, it’s unlikely your CDP implementation will completely address their goals. Worse, it won’t move the needle on your core KPIs.
All your customer data can be distilled down to two traits: who they are, and what they do. While we hinted at consistent naming conventions for events above, identity resolution is the other - much more challenging - side of the coin. If you want your CDP to live up to its promise of being the single source of truth for all your customer data, you need to clarify how all of your tools keep track of all your customers.
This also ties in with how each of your tools connects to the CDP. Only by working together can you get ahead of vendor complications and address any odd nuances involved in connecting various tools together. Because if there's a problem with your CDP data or its implementation, that’s going to hit every single tool downstream. And that means everyone - including members of your engineering, marketing, product, and sales teams - needs to have a seat at the table before implementation.
With a careful implementation of the correct CDP, it becomes much easier to address issues like these:
But before you can answer these questions, and long before you implement a customer data platform, it’s worthwhile to identify common pitfalls associated with a new CDP.
Evaluating, implementing, and onboarding a new CDP can be an excellent motivating force to restore data trust and fix problems with identity resolution in your tech stack. But it can’t do it alone. No technology can. Once you have a clear schema for naming conventions and a better understanding of your users, the right customer data platform can improve your lifecycle marketing efforts, add consistency to your sales pipeline, and accelerate digital transformation. However, if you start with inaccurate or incomplete data, the CDP will only exacerbate these issues. CDPs only began to appear in the early 2010s, but we’ve known about the challenges of ingesting poorly structured data for decades.
Consider the elements of an effective CDP strategy we outlined above. By implementing a CDP, you’re taking a big step towards becoming a truly data-centric company. On a deeper level, then, what you’re really doing is making it easier for every current and future employee to make data-driven decisions. This is why we constantly stress the importance of data trust, clear naming conventions, and identity resolution.
Managing your CDP must be an ongoing commitment because your customer data platform will, ideally, connect all your other tools for years to come. This points to another logical goal of your CDP strategy: consistent deployments of new tools. This way you’re not reinventing the wheel every time. The CDP is your wheel, and with each new tool you’re just adding a spoke and making it even stronger.
From another perspective, think of how your product roadmap might change over the next year, or how your marketing efforts will expand as you target new users. If you implement your CDP the right way the first time - and manage it based on consistent data governance principles - everything will be much easier from that point forward.
The whole point of a CDP is to centralize and organize your data. And it’s vital that members from all of your teams understand how the CDP works. But at the same time, inviting everyone to collaborate on this is not an invitation for everyone to create their own events. This can lead to duplication and confusion. We generally recommend companies track no more than 100 events. Any more than that, and complications can increase exponentially.
And while letting everyone create events can lead to confusion during CDP implementation and beyond, insisting that everyone in the company follow the exact same schema can slow down teams and reduce efficiency. So, where’s the middle ground? It’s possible to find that sweet spot where data shared cross-functionally follows specific naming conventions, while specific teams or groups have a bit of leeway when naming events that only they will use. We’ve seen this type of situation play out at Calendly, among several other clients. For a deeper perspective on this aspect of data governance, listen to our podcast episode here.
Even if it’s clear by now, it bears repeating: implementing a customer data platform could be the most cross-functional endeavor your team will ever undertake. We mentioned above how teams can run into trouble by keeping data and functions siloed within departments and workflows. And it’s especially apparent when it comes to CDPs. While implementing a CDP should be a cross-functional effort, the big irony is that engineers are often the best-suited to build it, even though marketing, product, and analytics teams will rely on it the most.
Failing to recognize this disconnect might just be the biggest pitfall on this list.
Engineers need to agree on naming conventions and fixes for identity resolution with all other teams before implementation begins. Nobody’s perfect, and no implementation is perfectly seamless. If you need to ask your engineering team to make fixes to the CDP after implementation, it happens. But everyone should focus on ironing out as many data issues as possible before the CDP goes live. That will make it much easier for everyone in your company to self-serve the information they need. Plus, it will go a long way towards demonstrating the value of the customer data platform to your engineering team.
AdvisoryCloud connects executives with tools and resources to monetize their domain expertise. AdvisoryCloud’s network of 11,000 professionals offers mentorship on a wide range of skills to the next generation of business leaders. But in order to grow their own business, AdvisoryCloud needed a better way to manage their customer data.
When AdvisoryCloud reached out to Mammoth Growth, their data collection strategy consisted of a traditional spreadsheet. This meant they were heavily reliant on manual reporting. Since there was no infrastructure in place to support any product analytics tools, Mammoth Growth’s first step was to build out the right data warehouse environment. Yet they couldn’t connect their data warehouse to anything else without a CDP.
Mammoth Growth helped implement a CDP from Segment based data governance and identity resolution best practices. Once Segment was in place, Mammoth Growth worked closely with AdvisoryCloud’s team to connect the CDP to all their different sources of customer data. This allowed them to streamline their operations and get automated reporting on their sales conversion rates for the first time.
Even if your data doesn’t live in a spreadsheet, there are many factors to consider when migrating from a legacy system to a modern data warehouse and CDP. And many of these details have their roots in good data governance. Please refer to this podcast episode for more details.
Nutrafol offers personalized, natural solutions for healthy hair growth through ecommerce solutions and partner referrals, and it is the 1 dermatologist-recommended hair growth supplement brand in the U.S. In 2019 Nutrafol realized many of their business processes were preventing them from reaching their growth goals. Like many other companies, Nutrafol maintained key reports in spreadsheets, and much of their customer data was siloed in multiple disconnected systems. Nutrafol turned to Mammoth Growth for guidance.
One of Nutrafol’s main challenges was effective data governance. In order to address this challenge in concert with their most important business questions, Mammoth Growth and Nutrafol decided to implement several new tools, including Mixpanel for product analytics and Segment as their CDP. Mammoth Growth advised Nutrafol that without a CDP, addressing data governance piecemeal across different platform integrations would have been insurmountable.
Once Segment was in place, Nutrafol was able to build rich user profiles and target specific audiences through their lifecycle marketing. For the first time, Nutrafol’s teams could access consistent data across their full tech stack with confidence. With organized customer data at the heart of their new tech stack, Nutrafol was able to grow revenue by 60% in 2020, which led to their acquisition by Unilever in 2022.
Glitch is a collaborative programming environment that helps anyone create powerful, easy-to-use apps. Glitch was growing rapidly, and to prepare for even stronger growth they decided to revisit their CDP implementation. Glitch relied on Segment for their CDP, and while Segment made it easy to create new events, organizing all that data fell to Glitch. They turned to Mammoth Growth for guidance in prioritizing which events mattered most.
Mammoth Growth helped Glitch develop a new, tightly-defined analytics schema, based on carefully structured data that aligned with their key business questions. Now that they’ve updated their implementation of Segment, Glitch has seen higher adoption of key performance metrics and reporting. This was only possible because their team members bought into Glitch’s vision of the CDP as their central source of truth, and then they made it a reality.
All of these case studies began with members of those teams asking questions like these:
Questions like these are normal and common. While every company will thrive with a slightly different data infrastructure, we’ve developed a simple, logical process so you can get the most value from your customer data.
In order to address companies’ goals for their customer data, Mammoth Growth developed our Analytics Architecture program. The objectives of this 8-10 week project are a complete audit of a company’s tech stack & data infrastructure, and a roadmap for improvements to their CDP strategy. Our Analytics Architecture program allows Mammoth Growth to develop a deeper understanding of the pain points and challenges each company faces with their customer data.
Due to the cross-functional nature of CDP implementations, we usually discuss updates to a company’s customer data strategies during any project related to data warehousing, product / behavioral analytics, and lifecycle marketing. Within each of these scenarios, Mammoth Growth follows these steps within our Analytics Architecture project:
If your company has built up legacy tech debt, there are additional steps in this process regarding how to manage old data before moving it into your CDP, or whether to abandon your old data and start fresh. While there are no easy answers for this, the goal of each Analytics Architecture project is a clear roadmap for how every company can accelerate their growth based on a solid data infrastructure. This project often signals the beginning of a company’s digital transformation, not its completion. Just like managing your CDP must be an ongoing commitment, true digital transformation is continuous.
For every company, there’s a unique combination of CDP strategy + technology + process improvements that can improve your lifecycle marketing efforts, add consistency to your sales pipeline, and accelerate digital transformation. And finding + implementing the right CDP for your team takes discipline and focus. Contact one of our experts today, and let’s talk about your customer data goals.