What is customer data debt? The simple definition is the amount of money required to fix data problems. Natural follow-up questions are what data problems, and how did they happen? As the word debt implies, the problems — lack of data quality, governance, security, repeatable processes and a lack of data standardization – can be caused by making trade-offs (“borrowing” data before you’ve “paid” for it), and these problems accrue over time.
Business users “borrow” data by using data from various sources without ensuring the data meets enterprise standards for quality, consistency, availability and security. This is like technical debt, where software may be released with design flaws or performance problems that later must be dealt with.
Like any debt, customer data debt is insidious in that it tends to spiral quickly, making it more and more difficult to get it under control, to identify the source and reverse the accrual. With customer data debt, that build-up also hinders the delivery of a personalized customer experience (CX). Marketers burdened with customer data debt essentially can’t trust that the customer view they’re working with is accurate, up-to-date or a precise representation of the customer.
Like an IT technical debt in which outdated or rushed software prevents applications from reaching their potential, having to “pay off” (fix) customer data debt uses valuable resources that might otherwise be spent making optimal use of incoming first-party data, building your own generative AI models or really just advancing a customer-centric strategy.
To Solve the Problem, Identity the Problem
To solve customer data debt, organizations must first recognize whether they indeed have it and, if so, recognize how pervasive it is. Observing and measuring the extent of the problem boils down to profiling your customer data to understand how fast it’s coming in, its cadence, and in what state; i.e., what has been done to the incoming data? Marketers need to understand where (or whether) data transformations, identity resolution, tuning, matching and the creation of a Golden Record have taken place.
Consider, for example, a situation where a company stores incoming customer data in a data lake. Perhaps a CDP vendor provides an assurance that incoming data has been matched. If you’re an operational marketer with multiple records for one customer and you need to know which of two different email addresses to use for a campaign, you may have to use one without knowing if it’s the right way to reach the customer. That’s customer data debt. There may be dozens of customer or transaction tables, tons of redundancy and a complete lack of consistency that prevent the delivery of a hyper-personalized CX.
To remedy the problem, the marketer has to know the source of the data, its recency and the context under which it was collected. Like a polluted river mouth, it’s easy to spot the detritus but you need to go upstream to determine the origin.
Customer Data Debt and a Composable CDP
Customer data debt is not a new phenomenon nor a new term, but it is becoming a larger part of the discussion around a personalized CX in large part because of the composable CDP trend.
Many CDP vendors that offer a composable architecture framework do not consider customer data debt the responsibility of the platform. That is, they will say that a modular CDP will integrate customer data from all sources and provide a single customer view, but with the assumption that the steps needed to prevent customer data debt from accumulating – cleansing, identity resolution, security, transformations, etc. – have either already been completed or will be outsourced for completion downstream.
In the example of the marketer with two email addresses, that “single view” might contain two (or more) identifiers for the same customer, but as we’ve seen a simple match does not make it actionable for the marketer. Will a database administrator clean up the transaction or customer tables? A third-party data transformation tool? Will that be done in real-time, or at the very least in the time needed to keep up with the cadence of the customer?
Solve Customer Data Debt with Redpoint
The Redpoint CDP composable architecture framework is unique in that Redpoint solves for customer data debt at the source. The platform performs all data quality, data transformations, data enhancements, identity resolution and the creation of the Golden Record as data is ingested.
Customers using one or all of the platform’s three main features – CDP, journey orchestration and real-time interaction – can choose any of the underlying services to support business use cases from ingesting data through to the delivery of content. Whatever the business use case, marketers and business users can trust that customer data has been made fit-for-purpose because the platform handles all the data cleansing tasks as data arrives. All customer data is ready for prime-time, in other words, the moment it is accessible.
Composable architecture applies object-oriented principles to software, but if assumptions are made about data, the danger is that any problems with the data will result in compounded customer data debt throughout the composable framework. The problems with data don’t even necessarily have to be problems, per se, but even different definitions. Perhaps one table uses a month/day/year field, while another uses year/month/day. Or in one source a data clerk is instructed to enter 0’s for data not captured, while in another source the clerk enters X’s.
In and of itself, those issues may not seem insurmountable, but that’s what’s so insidious about debt. Its presence requires attention but the organization may not even know it’s accumulating. Speed-to-value may be one of the stated benefits of a composable CDP, but customer data debt will negate any value until the debt is paid. Left unchecked, customer data will lead to an inferior CX. Your customers expect better, and they will go elsewhere if their expectation for a personalized, omnichannel CX is not met.
