As digital media grows more complex and more user data is created and collected on a daily basis, marketers, agencies and publishers need better ways to buy, sell and manage audience data beyond the standard suite of analytics tools. How do you capture all this valuable audience data, shape it into insights and then activate it to drive outcomes?
The answer is with a DMP, or data management platform. We’ll get to the particulars in a bit, but a DMP collects, stores and sorts digital information. It provides useful snippets of data and information marketers, publishers and businesses can use to their advantage.
You may hear the phrases “big data” and “data management platform” used interchangeably — because big data and DMP are one and the same.
Unfortunately, although big data and user data are instrumental, you can’t do much with the raw information. You need it sorted and converted into a usable form, at least so you can understand what you’re looking at. This is the power of a DMP.
Even knowing all this, there are plenty of questions that come to mind. What is data management, for instance? What does DMP mean? How does a DMP work? What does a DMP do?
Huge collections of data need to be stored somewhere. Not just that, they need to be organized, sorted, translated and shared with the appropriate parties. You can’t just take data, throw it into an empty box and hope something comes from it, as wonderfully convenient as that sounds.
A DMP is the backbone of data-driven marketing and serves as a unifying platform to collect, organize and activate your first- and third-party audience data from any source, including online, offline or mobile. Essentially, it is the data warehouse where the data is not just collected, but facilitated.
A DMP not only collects and organizes your data, but makes it available to other platforms (such as DSPs, SSPs, and ad exchanges) to be used for targeted advertising, content customization and beyond. Some people describe a data management platform as the “pipes” of ad tech — connecting many platforms in a neutral way so marketers can use their powerful audience data when and where they want.
The question “what is a DMP” is black and white, but the answer isn’t so simple. It’s not just a place to store information of all kinds. It’s all a tool or piece of software that makes said data practical. Having an endless supply of data without a DMP is like having a key without a lock.
Sure, with the raw data you can solve a problem or two, but ultimately, you’re not getting anywhere.
A DMP can collect unstructured audience data from any source, including desktop, mobile web, mobile app, web analytic tools, CRM, point of sale, social, online video, offline and even TV. A viable DMP should collect audience data on more than a surface level, going far beyond things like URL and keyword information.
This first-party data — in other words, the data you own, not purchased from someone else — can be collected based on specific behaviors such as clicks, downloads, video uploads or video completions, interests like sports, football, parenting, museums and travel or demographic information. Therefore, we can conclude the true meaning of data management is to prepare data, of any kind.
Naturally, the DMP must have direct access to these data sources. This also explains why data professionals often recommend integrating a DMP into your work and processes from the start. It’s much more difficult to feed existing data into a tool, rather than letting the tool collect this information on its own.
Of course, this is all very generalized. If you’re not already familiar with DMPs, you may have no idea what kind of information is usually collected or processed by these platforms.
Lotame’s data management platform collects audience data from the following online and offline sources:
Once the first-party data is collected, it is organized hierarchically into a series of segments, which can change based on each end user’s business models. A large publisher network may have their hierarchy divided up into different buckets based on each of the individual websites they own. An agency can have separate accounts for each of their advertiser clients.
All of this data, once categorized, can be used to better understand your audience, create effective RFP responses, enrich your audience to learn more about them and extend your audience reach to address campaign commitments.
Think of a DMP like a mail-sorting system, only for digital content and data. As information comes in, it is fed to the appropriate department or category for the related team to work with.
To fully understand the benefits of a DMP, you must break the collection and organization processes down in more detail.
A DMP will organize your first-party audience data into categories and taxonomies, which are specified by those using the platform — in this case, that would be you. You define how that data is organized, which means you need to understand — and define — what you need out of your data before deploying a DMP.
Most of the time, the organization process is way more complicated than simply layering data into base taxonomies and categories. This is due to the nature of your audience. A clothing retailer, for instance, may have several demographics they want to understand, from teens to adults. Then there’s gender, age and much more.
You can simplify this process both for the DMP and yourself by creating audience segments, classifying parts of your audience as a like-minded or similar group — for example, all your desktop website users versus all of your mobile app users.
Marketers, publishers and advertisers frequently rely on audience segmenting but it’s certainly not exclusive to those groups.
Soon after the data has been organized and classified, you can take chunks and analyze it to discern customer patterns, trends and intent.
This is where you work to understand all of the data and information you’ve collected. The process is made much easier thanks to the categorical layout and meta-details the DMP has sorted your information neatly into.
The final step is to transfer, activate or use and scale the data. As with some of the other steps, how and what you do with the organized data relies on the industry you work in. Advertisers and marketers, for instance, are not going to use data in the same way as a publisher. That is true even with identical sets of data.
No team or department will use data solely for the same purpose. That means the data context must be decided beforehand so you can select which data goes where in the sorting process.
Once you’ve collected and organized your data, it’s time to build audiences. For any particular targeted campaign, a marketer can predefine the target audience for that campaign. So instead of choosing to spend your advertising dollars on a particular financial website to reach an audience of folks interested in finance, you can instead build a financial audience based on actual behaviors across the wider Internet.
Audiences can be built inside the data management platform using any combination of first-, second- and third-party data sources, which are combined using Boolean logic, which breaks down all values into either true or false. You can build as niche an audience as you wish, and are only really limited by the scale you need. Generally, the more niche the audience, the smaller the size.
Hands down, you’ll know the best way to classify your audience and related data. If you’ve been in business for years — decades, even — you probably already know your audience, demographic and targets. Keep that in mind when looking over this next section. Even though we may point out some elements to consider, the list is not comprehensive and may differ for you and your needs.
Here are just a few of the data points you can use to build your audience:
You’ll want to consider what you’re collecting the data for and how you want it separated before defining data points, obviously. This can seem daunting, but it’s quite easy once you dive in.
You must also consider the data sources. Is your data coming in from first, second or third parties, and does it need to be separated and classified as such? If it does need to be separated, you’ll want to come up with an additional set of taxonomies for each source.
As with most tools, a many types of professionals and businesses can benefit from a DMP. Upper or C-level management might rely on a DMP to make smarter and more informed decisions. IT or networking professionals might rely on a DMP to maintain and operate a companywide system, gaining insights from the tool that can be used to choose machines, software and more.
The most common parties using DMPs, however, are publishers, marketers and agencies.
Publishers use data management platforms to manage the audience data collected from all of their websites and the data-driven advertising campaigns run across those sites. A DMP enables publishers to capture first-party audience data and enrich it with additional audience insights, allowing the publisher to increase CPMs for both direct sold and programmatic inventory.
Marketers and agencies use data management platforms to identify and classify audiences at a significantly deeper level and to gather an extra layer of data about their audience, regardless of the data source. These insights into your most valuable customers allow marketers to identify and target prospects who look and act exactly like them so you can increase your audience base.
At any given time, marketers are sitting on a heaping pile of data. Without the technology or tools to make it readable, all that data is useless. Naturally, they should be more concerned about how DMPs work, and what a DMP does with the data, than whether or not they should use one.
Marketing teams should always use a DMP. But how do you implement one into your marketing strategy? Do you have to change anything?
The good news is, you shouldn’t have to change a thing. Of course, that depends on whether you choose a good DMP. With the right platform, you should be able to use your existing data, profile your existing customers and audience, extract meaning from your data — old and new — and, finally, improve existing campaigns or come up with new ones altogether.
A marketer’s mission is to predict customer behaviors and use datasets to make proactive moves or decisions. What advertising and marketing materials will resonate with your audience most? What content will get the most engagement from your customers and audience? What message will inspire them to buy your products or pay for your services?
A DMP should be the core of your marketing processes and strategies. Use it to collect any and all data, and always reference it before making a decision or taking action. Think of a DMP as the last line of defense in a battle. Before rushing onto the field, weapon in hand, you stand your ground and consider the best course of action.
DMPs offer users many ways to activate data on any channel, including:
More important, however, is what you can do with all that data you collect. Here are just a few things you can focus on and what you can do with them:
Out of all those scenarios — especially in e-commerce and marketing — delivering a personalized experience to your customers is one of the most important. You build a rapport or relationship with your customers by focusing on their interests and needs. Using data to personalize their experience is a great way to do that. It’s why brands like Amazon offer “recommended” sections with products similar to what a customer has viewed in the past.
Thus, you can see the value and importance of a DMP. Without one, you cannot offer the kind of experiences being discussed above.
Look, we’ll level with you. Chances are, you’re already sitting on a mountain of valuable data. Most people understand the importance of collecting data and storing it for later. Unfortunately, storing and saving data is different from classifying and using it.
A DMP can help you turn that mountain into a gold mine! To think, it’s sitting there right under your nose waiting to be put to good use!
So, now that you’ve read this whole article, we’ll ask you: What is a data management platform?
If you don’t feel confident answering that question, don’t worry. Lotame would love to walk you through and explain everything. Contact us, and we’ll be happy to give you a demo.
Updated on September 19, 2017