Identity Management: The Digital Marketer’s Swiss Army Knife

Identity Management: The Digital Marketer’s Swiss Army Knife
Adobe Products Featured

How many connected devices do you own? For most of you, the answer will be more than one.

Now consider how consistent your experience is with any given brand across these devices. Only a few brands come to my mind that have been able to speak to me with the same message consistently across all the devices I own.

Being able to effectively manage and measure a cross-device experience in full accordance with data privacy and marketing laws is a growing problem for marketers and a huge opportunity for digital marketing. As a result, it’s an area where the concept of identity management has flourished.

The most popular application of identity management in digital marketing has been the device graph (see here for the definition of a device graph). It’s often used by marketers to increase campaign reach without sacrificing audience accuracy. To do this, the marketer will define a selection of devices to target with a campaign. Then, rather than targeting this group of devices in isolation, an option is selected in the UI to expand the reach of the campaign to target all the devices connected to the original set, based on the mappings of a device graph.

Which device graph you choose to use is completely dependent on the technology in which you choose to define your audience and to what data that technology has access. For example, in Google’s Doubleclick Bid Manager, you could use Doubleclick’s device graph, whereas in Facebook’s Power Editor, you could use Facebook’s device graph and in the Adobe Experience Cloud you could use the Co-op device graph (see how many devices the Co-op device graph has connected to this browser here). The expected outcome is an increase in the size of the segment.

That said, this binary approach of “flipping a switch” to increase reach only scrapes the surface of what is possible when you can identify, collect, and merge profile information at the device, person, household, or even business level. Only with this level of flexibility and control can you create a truly meaningful experience for a given audience.

To showcase the full extent of this opportunity, I’ll outline three examples of how identity management can be used to better target audiences in your marketing campaigns. Think of these as three tools in your “digital marketing Swiss army knife:”

1. Understanding cross-device user journeys — As a marketer, I understand that a user journey can transverse multiple devices. I don’t want to limit the size of my audience by evaluating each device’s attributes individually if that device can be linked to others owned by the same user, which may have seen the attributes I’m looking for.

Consider the example below. Here I have three device profiles tied to the same person by a common ID, the Device Graph ID. I want to define an audience that has viewed the flights, hotel, and tours pages of my website as I consider them in market for a package holiday deal. I know that if I merge all three device profiles, the superset of attributes qualifies for my audience segment. I also know that if this person belongs to my audience based on their activity across all three of these devices, then all three of their devices should be a part of my audience also.

The result is that I’m able to qualify and target all three devices as being part of my audience segment because their merged profile qualifies for that audience segment.

2. Audience targeting on a shared device — As a marketer, I want the ability to be able to target an individual, not a device. This is critical when the device is shared across multiple people and the message I need to deliver is specific to an individual based on their user profile.

Consider the example below. Here I need to be able to define an audience based on a user-level attribute stored against a known identifier (credit score > 700). I only want to target an audience based on this attribute when the shared device is being used by people who have this attribute stored in their profile. This can only be determined when users are in an authenticated state (logged in) on a given website or app. I do not want to evaluate the attributes associated to the shared device since they could be associated to anyone using that device in an anonymous state.

The result is that I’m able to qualify only Person 3 for the segment by choosing to evaluate their profile separately from the profile of the shared device (which Person 1, Person 2, and Person 3 had been using).

3. Targeting by identity — As a marketer, I’ll want to define audiences based on a combination of data sets stored at different levels of identity. Typically, these levels range from the anonymous device to the person, household, or business. To be able to combine these data sets, I’ll need to ensure that the system I use to define the audience can evaluate the attributes I choose, based on the level of identity to which they belong, and that I can target the entire audience at the correct level of identity.

Consider the example below. My audience definition is all households with an income greater than $100K per year that own an iPhone 7 on Data Plan B. If I were to evaluate the attributes without considering the level of identity each attribute belongs to, both Household 1 and Household 2 would be part of my audience. Once I have correctly selected Household 2 as qualifying for my audience, I’ll want to ensure that only the one device in the household that meets my device-level requirements is qualified for the segment.

The result is that I’m able to qualify the specific device and household which meets my audience segment definition and I’m able to send these qualified devices downstream for targeting in their specified platforms.

To execute on any of these use-cases requires the capability to merge profiles, evaluate a super-set of segments, and target multiple devices in real-time. To do this, you’ll need to invest in a data management platform (DMP) capable of storing and dynamically merging profile data at the device, person, or household level. Adobe Audience Manager is a great example of a DMP with this capability.

Click here to find out more.

Recommended Articles