First things first: what is a DMP?
In the last decade, Data Management Platforms, or so-called DMPs, have been at the center of the programmatic world. Mostly used by advertisers and publishers, DMPs are widely used for collecting and managing user data in an anonymous way.
DMPs’ first job is to collect the data from a variety of sources: website, mobile app, digital advertising, CRM, etc. Once in the DMP, that data is stored and analyzed by the software. The goal is mapping all that data together into an anonymous customer profile.
When the data is ingested, it can be pushed to any addressable advertising platforms available as a destination in the DMP. This can be to personalize your website experience or to target users in paid media campaigns.
DMPs differ from CDP in many different ways. You can read more about what a CDP is in our article “What is a Customer Data Platform?”
DMPs aren’t dead just yet!
In the recent years, especially after Google’s announcement to retire 3rd party cookies from their environment, DMPs have been condemned to die. It is true that the post-cookie world as well as the many new consumer privacy laws and industry-changes have put a lot of pressure on DMP.
In fact, their core historical identity cannot survive in today’s world. But that doesn’t mean they can’t evolve and survive. In a market where consumer privacy has become the norm, DMPs simply need to adapt – just like others.
It is important to keep in mind what DMPs were built to do, not how they operate and how the pipe connect per se. The essence of a DMP is to enable a brand to capture, enrich and activate its customer data in one place. Today, this heavily relies on 3rd party data indeed, but tomorrow, it may very much rely on 1st party data only.
The question remains though if we’ll see new technology, like CDP, emerge to go on top of DMP to securely collect PII data or if that functionality can be built in directly into the DMP.
DMP use cases are still highly relevant today
Thanks to its marketplace, DMPs typically offer the ability to enrich 1st party data with 3rd party data. In other words, it allows you to create a super segment to portray a better picture of your audiences.
Let’s say you want to target your website visitors. You don’t know their location, gender, or income level – there are many 3rd-party data providers that could provide you with that info. And trust me, 3rd party data comes in all forms and shapes. For instance, a data vendor could provide you with data about your customers regarding purchase intent, location, etc.
Paid media Activation
Your DMP can help reach your customers and prospect in the outside world with scale and efficiency.
On the one end, you can activate your 1st party audiences in any addressable media-buying platform that can synch with your DMP (native or custom connections). In practice, you can serve targeted ads to users who may not engage with your emails or cart abandoners.
On the flip side, you can also use your DMP to efficiently augment the reach of your paid media activation. From DMP audiences, you can run lookalike modeling using your 1st party data as a seed audience.
One thing to clarify about lookalike is that they are not solely relying on cookie data. For instance, in Facebook, you can easily run a look alike from your first party data using Facebook proprietary audience data. This will result in highly qualitative propensity model to find people who look just like your best customers or prospects.
So far, we’ve talked about activating your audiences. A DMP also allows you to suppress some of your users from specific activation.
The benefit of this you asked? Well, think about that conversion campaign you are running with the objectives of selling your latest pair of sneakers. You may want to suppress recent buyers from targeting. These probably won’t be buying a new pair of shoes for a few months or so.
As easy as it is, suppression use cases make your media activation more efficient by not targeting users who are unlikely to be receptive or to convert from the start.
But do I really need a DMP?
For any addition to your tech stack, you should always do a Return on Investment (ROI) calculation. Doing this may help you understand whether the added-value will be worth the investment. That is a highly mandatory step after validating that the tool is a fit for your organization. You should check this from a business objectives, team, processes, and existing technology perspective.
One exercise I like to do with my clients is asking those very simple questions: what…
- is your current media mix, and do you rely heavily on addressable media like programmatic or direct publishers?
- is the overall addressable media budget?
- use cases do you plan on using your DMP for?
- is the size of your 1st party audience?
The answer to those four key questions will help you understand better the potential ROI that you will get on your investment.
What your answers might mean
|Media plan||If 75% of your activation is through addressable channels, many DMP use cases may be applicable to your business. On the other end, if you only do run on traditional media, it will not be worth the effort as the DMP will obviously not operate in those environments.|
|Media budget||If your media plan is 80% on addressable channels but your media budget represents a fraction of the cost of DMP, this is likely not the right investment for your business.|
|Audience size||If you are managing an ecommerce platform capturing thousands of visitors or transactions per day, chances are you will see lots of benefits in (re)activating them as you have scale. However, if you only see a thousand visitors per month and do not have any other transactional data at your disposal, the use cases you can deploy with your DMP will be highly limited – hence not worth it.|
|Use cases||If you have scale and budget that justifies the investment, it is also important to think through the use cases for the technology. Will you gain enough in media efficiency to justify it? If you only activate a small subset of your site visitors but continue to buy 99% of your inventory directly in the DSP, it would be limited in value.|
DMPs – not all but some – are catching up with the recent industry and privacy changes. They evolve, they adapted, they are still worth considering depending on your business’ ambitions. There are still plenty of relevant use cases a DMP can help a business with. Some examples are audience enrichment, paid media activation, personalisation or even suppression.
Yet, there are two main questions. The first one being: “what are you trying to do“. The second question is: “is it a fit for your organization from a scale, budget, and objectives point of view”.
On the other end, if a DMP is not a fit for your organisation at this time as you may mostly work in known media channels (ie: email; social media, etc) or have a large database of 1st party data. Chances are a CDP may be a better fit for you. You can read some of the use cases a CDP can quickly unlock in our article “Three Simple Strategy to Quickly Win with your CDP”
Don’t hesitate to reach out if this is a question you’d like us to help you answer!
About the Author: Elaine Lorent
Elaine is our Head of Data Strategy at NOBI. She spent 8+ years in Media and Digital Analytics, with focus on CDP, DMP and CRM strategies. Elaine has a Master of Science In Integrated Marketing & Analytics from the New York University. Her passion is to help brands leverage their 1st, 2nd & 3rd party data in a meaningful and responsible way, to become fully data-driven organisations.