These days, consumer data is more important than ever, and enterprises are increasingly turning to Customer Data Platforms (CDPs) to help them make sense of the mountains of data they have at their disposal. But not all CDPs are created equal — especially when it comes to timing. Many CDPs only provide "near" real-time data, which is a problem for companies that need to make decisions in the blink of an eye.
Chances are, if you're reading this, you're already familiar with the concept of a CDP. But in case you're not, a CDP is essentially a software platform that pulls an enterprise’s customer data together from multiple sources into a single record. The data can come from a variety of sources, including first-party data from company-owned properties like websites and apps, second-party data from business partners, and third-party data from public databases.
Now that we're all on the same page, let's talk about real-time data. True real-time data (or “live-time data”) is, quite simply, data that’s updated as soon as any changes occur. This is in contrast to batch processing, which is when data is "updated at predetermined intervals." So, what does this have to do with CDPs? As our friends at Teradata put it, “A Customer Data Platform (CDP) is a type of packaged software which creates a persistent, unified customer database that is accessible to other systems. Data is pulled from multiple sources, cleaned, and combined to create a single customer profile. This structured data is then made available to other marketing systems.” This is key — because the process of pulling, cleaning, and combining can be instant, or take days. And how fast it all happens determines when it’s available to your other marketing systems.
Near real-time data is data that’s delivered within a few seconds of being generated. This may sound fast, but for companies that need to make split-second decisions, a few seconds can mean the difference between success and failure.
According to Validio, “Real-time processing usually takes place in sub-milliseconds to seconds, while near real-time processing takes place in sub-seconds to hours. This means that real-time data processing is much faster than near real-time processing.” The problem is that when it comes to data capture and CDPs, every vendor will tell you they’re giving you real-time data, when in fact it’s near real-time. Or the data is captured in real time but takes hours or days to deliver for activation. Since the term “real-time” has been heavily misused in the industry, for clarity we’re going to refer to near real-time as real-time data, and true real-time as live-time data.
Analysts (Forrester, Gartner, CDP Institute, etc.) are always talking about how 90% of all CDP projects fail. Why? Because CDPs bring data together but rely on the tech sending them the data to solve all their data issues, which they don’t. So, you end up with incomplete and inaccurate data from a variety of sources and spend all your time trying to use it — instead of actually using it. CDPs aren’t generally live-time because they’re dependent on the source data and the time it takes to get that data together, clean it, and structure it for use. You end up with a massive dataset that isn’t available when you need it, isn’t accurate, doesn’t solve today’s digital challenges, and doesn’t like to play well with others (i.e., integrate with your other systems).
To provide the most accurate and up-to-date information possible, a CDP must be able to process data in live time. But most CDPs on the market aren’t capable of live-time processing because they rely on batch processing to update their records. This causes significant delays between when a change occurs and when that change is reflected in the CDP's database. And we all know when it comes to data, time is of the essence.
Take marketing campaigns as an example. Let's say you're running a promotion and you want to target visitors who put something in their cart but didn’t check out. Would you rather target them in-the-moment, while they’re actively on the site? Or wait until the next day when they may have moved on to something else or purchased the item from a competitor? While one day may not seem significant, it can be the difference between a conversion and a lost sale. And, when you’re talking about tens of thousands of purchases a day — the impact is staggering. On the other hand, if your CDP is updated in live time, you know you're always targeting your offers based on the most recent data possible.
A modern, live-time data solution can either enhance or replace your CDP, delivering data in live time to maximize efficiency. At the same time, it solves digital challenges like identity for anonymous visitors, enables multiple integrations for activation, and manages compliance throughout.
To understand if your CDP is providing live-time data, ask your provider how often their platform ingests new data. If they say anything other than "in milliseconds," you're not getting live-time data.
Another option is to run a quick test. Simply generate some test data and see how quickly it appears in your CDP interface. If it takes a few seconds (or more), you're not dealing with a live-time platform.
Finally, you can look at the latency statistic in your CDP's reporting interface. This number will tell you how long it takes for new data to be processed and made available in the platform. How quickly the data’s available to use is the key to decisioning and activating it for use in marketing, personalization, and decision making. If this number is anything other than 0, your CDP isn’t live time.
It's more important than ever to have access to accurate, up-to-date information. CDPs are a valuable tool for businesses that want to consolidate their customer data into a single, comprehensive record. When it comes to choosing a CDP for your business, don't settle for anything less than true real-time (i.e., “live-time”) processing and activation. Your business — and your bottom line — will thank you.