I’ve been an insider in understanding how to manage and account for the notion of “identity” on the internet for more than 5 years.
I can tell you that identity resolution as a tool for marketers is at once both a pretty simple idea and complex in scope. And if you apply it to your business, it can be a profoundly powerful tool.
It’s simple: we want to understand that the phone in my pocket, the tablet in my bag, and the PC on my desk are all the same person. So that we can market thoughtfully to people instead of haphazardly to the billions of devices in the world.
It starts and ends with a simple question. Who is this person?
When wielded fully and thoughtfully, identity can change the way we do business entirely. When dropped into a business with a little forethought, it can languish and be seen as a huge waste of time.
I’ve seen both types of implementations over the years. Inevitably, those that think through their plans and take the time to integrate identity resolution into all of their processes are extremely happy. Often those who shoehorned in identity resolution as a point solution were not.
The challenge architecturally is that identity resolution is a service that drives other services. Isn’t something that stands on its own. It’s information that you can use or not use for various tasks within your business.
It’s information about information. And because of the “meta” nature of it, I became passionate about identity resolution as soon as I understood it.
Gold Rush and Identity Resolution
I grew up in the San Francisco Bay Area, which was the epicenter of the 1849 California gold rush. Interestingly, as kids, it’s not really the 300,000 plus miners that flocked to California that we study in school (though a few of them literally “struck gold”).
The story of the gold rush is told primarily through the stories of men like Levi Strauss, the man who made a fortune selling blue jeans. Also, Samuel Brannan, whose fortune was built on the sale of pick-axes and shovels.
The most valuable thing in California in 1850 wasn’t gold. It was the tools needed to find gold.
I view identity resolution in a similar way. The growth in the digital economy, and in particular the App economy in the past 10 years feels like a gold rush.
Certainly, some digital properties are going to strike gold (Uber, Yelp, Amazon). But the vast majority are not, but the big players that tell the story of the digital gold rush are the enabling technologies.
Even now, in the Bay Area, we hear much more about Larry Ellison (Oracle Executive Chairman). Mark Benioff (Salesforce CEO), and Jeff Lawson (Twillio CEO) than we do about Travis Kalanick (Uber founder) or Jeremey Stoppelman (Yelp CEO).
With a dream of influencing the larger history of this digital gold rush, I’m excited to help people acquire helpful tools. Also to know how to wield them, so that we can all succeed in finding digital gold.
How to succeed with Identity Resolution for your business
I talk with clients about Identity systems regularly and often get asked how to set them up to be the most effective.
The short answer is you have to use it, and use it everywhere. The long answer is that using it everywhere means it must be an internal API-based service that all of your other existing systems integrate with and leverage.
If your team monitors the ad dollars you spend to deliver eCommerce conversions, then this system must query for Identity each time a conversion happens. And also see if any of the IDs that represent that user was advertised to.
If you are trying to reach users in a certain demographic or those who have been to specific locations. Then you need to build a list of those people based on the device that took the action, then extend that segment to all of those people’s devices.
On the other hand, if you just sold someone a sweater, you need to make sure your blacklisting tools don’t target (or retarget) them on any of their other devices.
When you should use identity resolution?
The list goes on, but the idea holds… Before you take action on the information you’ve developed or acquired about a person, your first step should be to ask “Who is this person?” In a digital sense so that the action you take applies to all of the places you might find that person.
My favorite example of the “who is this person” question being asked particularly well is with a retailer I worked with some time ago. Whose number one marketing strategy was email remarketing.
In general, their email marketing worked pretty well, but they didn’t have email addresses tied to a number of the devices they saw.
They built an identity resolution toolset that included their own knowledge of consumers as well as that of multiple third-party vendors. Then asked the third-party vendors to map the unknown IDs to other, known IDs that were tied to email addresses in their system.
Then they updated their automated email system (which at the time just disregarded pageviews not tied to an email) to ask the question, “Who is this person?” To their internal service and email the appropriate address if they got a match via their service.
The resulting uptick in emails that drove conversions was worth millions of dollars per year.
That’s the tough part. But it is doable, trust me!
A little technical context of identity resolution (but don’t worry it’s not that bad)
I like to think of an Identity graph a bit like an org chart for a company. There are 2 Major Identifiers, the Household ID (HHID) and the Person ID (PID), and many other minor IDs. (cookie/mobile device id/CTV ID/email address/address/phone number/etc.).
The PIDs are the “direct reports” of the HHID, as are characteristics that are better served to map a household (like Address or CTV ID). So a HHID has 2 or 3 or more PIDs and each PID has several ways IDs are tied to it.
Each of those ways to map creates a list of IDs that are connected to the PID and HHID
After you outline the general org chart structure there is a decision to make: Either you must keep track of the sources that indicate that a specific Identifier maps to a PID or HHID based on multiple data sets and place it only in the highest confidence data set, or you must add that ID in multiple places.
I recommend the latter for ease of use in querying the graph later, but it’s a subject on which reasonable minds disagree.
The next step is to build an API that allows all your various other services. In order to help them understand “who is this person” and get a response about their interests and IDs that are tied to them:
Then, every action you take, from site personalization to off-site targeting first asks our critical question: “Who is this person?”
By asking yourself this simple question repeatedly and looking for answers with a little help from an adtech partner, you can have a profound impact on the growth of your business.
About the author
Kevin Mullen is the Managing Director of Roq.ad, a European-originated Privacy-by-Design AdTech company. His career has always been at the intersection of Mobile, Big Data and Analytics.