Consider a typical family household. Would you want the same TV ad targeted to the college kid as to the fifty-something dad? In all likelihood, you would not. That would be a waste of ad spend and additionally, irrelevant and perhaps annoying to either the dad or the kid. Instead, you’d want to target each consumer individually—and be able to determine, with the help of identity graphs, when each individual was watching his or her show and what ad would best appeal to them in that moment. While household data might be useful for some types of consumer targeting, creating identity graphs and focusing on the individual as opposed to the household is often the way to go.
Digitally, a household is connected through and through. Family members all stream on one network, using and sharing multiple devices depending on where they are in the house—be it laptops, phones, connected TVs (CTV), or connected appliances like smart fridges. An individual might start browsing for a show to watch on their phone and then choose to watch it in the family room—or a parent might look for a new product to buy on the family iPad and fulfill the purchase on their personal laptop. Altogether, everyone in a household contributes to a specific digital environment based on how and what devices they browse on. This forms a unique household data fingerprint—which, in some respects, seems easy to create targeted advertisements for.
Household data can be one avenue of reaching an audience, especially when the product in question would be best targeted to more than one individual in a household. Consider home insurance policies, meal kit companies and streaming service subscriptions—oftentimes, one subscription per household is more than enough, and ads aimed at each household member can be redundant. Or think of the purchase of a new family minivan. Family members form a circle of influence, and the desires of all family members can serve a significant role in this type of financial decision making. Thus, marketing to the whole family—such as through a target TV ad—makes more sense in this case than marketing to a single individual in the house.
However, household data can’t serve as a stand-in for individual consumer identity data. You wouldn’t serve the same ad to the college kid spending winter break at home and to her parents, who likely have remarkably different interests and styles of life. Just because you’re getting a ton of information from one household doesn’t necessarily mean it’s relevant to everyone residing there. Additionally, depending on your campaign, not all data collected in the household is meaningful or relevant data; knowing what data matters for what situation is just as important as getting the data in the first place. Whether marketers should group individuals under a single household or treat each consumer is based on the product at hand, and in many cases, constructing individual identities for each consumer is the way to go.
Brands today rely on much more than just an undivided collection of data for each household. They construct identities for each individual to best target and create personalized ads for each consumer. By tracking a user across multiple devices and building an identity graph, marketers can be more certain of when a specific user is using a device and what their interests are. By marketing to the individual, there are fewer wasted viewable ad impressions and there’s a higher likelihood of clickthroughs and engagements.
Use identity data to deliver relevant ads
Roommates might live together, but not shop together. They might use the same network to find recipes and plan their grocery shopping, but they likely buy their own groceries and make their own meals. Two roommates are probably not making significant joint purchases and will typically have different viewing preferences and different interests, so it would hardly make sense for them to be targeted with the same kind of ad. The household data fingerprint for roommates represents two or more eclectic tastes between individuals who are best targeted individually than as one whole unit.
Also, in the case in which it is a family unit living in a family home, it’s not always just family in the house—visitors come and go. Friends and relatives might visit for a few hours, and friends from out-of-town might stay a couple of days or longer. Children come back from college for break, often followed by friends stopping by. Plenty of people other than family are connecting to the network, and making changes to the household data fingerprint.
All of this can create irrelevant information for marketers. Those in a house could be a group of disparate individuals who cohabitate but maintain their own wildly unique interests, diets, and personalities. In the case of a family home, lots of individuals—some not even members of the household—end up connected on the same network, slightly altering the digital fingerprint of the household with their personal online activities.
Two roommates might share the same network, but one of them might be vegan and interested in organic produce, while the other might be on a quest to find their city’s best barbecue. A household mainly inhabited by two 50-somethings might be interested in an advertisement for Billy Joel concert tickets. But when their child comes home from college, their searches, views, and likes may tilt the scales in the direction of a festival headlined by Billie Eilish. In these cases, consumer targeting built on household data can be a bit unreliable and send mixed messages to the audience. Rather than wasting impressions on targeting these groups as a whole, it makes more sense to target the individuals personally.
This is where identity-driven marketing and identity graphs become extremely valuable. Rather than viewing a household as one individual, it’s crucial to look at each person in a household separately. By tracking an individual’s interactions with a website or product across a set of devices and identifiers, you can build an identity graph.
Advertisers now can collect online data that links various data points derived from activation. Device data, location data, consumption data, unique identifiers and IP addresses can now be stitched and linked back to offline identifiers like postal addresses. Even when a user uses multiple devices (as many consumers do these days), an identity graph can track who they are and what devices they use based on factors like what websites they browse on, IP address, and location throughout the day. These links create a way to then connect offline data (demographics) to online data (devices) to create an ever growing, and stronger, identity graph.
What is an identity graph? – An identity graph is a database that collects all known identifiers of an individual customer. This can include personal identifiers like phone numbers, email, account usernames, and even offline data like a physical address. The identity graph collects these identifiers and connects them to a customer’s profile. These particular identifiers, including account logins and user profiles, make up deterministic matching—accurate data traceable to one individual. However, what deterministic matching lacks in scale (not all websites require a user to login or input an email address), probabilistic matching makes up with its wider scope. Probabilistic matching, though not 100% accurate like deterministic matching, can track interactions across multiple devices by looking at browsing activity, IP address, location, and purchase history. An identity graph uses both deterministic and probabilistic matching (identifiers and interactions) to track an individual across multiple devices.
It’s critical to know where to look for the gaps in your existing data and understand how to identify targets as they move across devices and locations throughout the day, as well as the platforms that can link them in a scalable way. One of the best ways to fill in these gaps is by grouping together both first- and second-party data and synthesizing billions of online interactions with multiple identifiers. It’s key to develop dynamic profiles for each person within a household, as opposed to mistakenly looking at everyone in the household as “one person.”
Individual consumer identity data is a powerful tool. Accurate consumer identity data improves customer experience and increases revenue. An identity graph is a complete picture of a user, and it tends to be comprehensive and stable across devices. The college kid can be served the festival ad, while his parents might see an ad for a Billy Joel concert, all in the same household.
Move the right person along the funnel
Another place where being able to better link individuals to devices within a network is no longer having to take the “spray and pray” approach with your marketing messages. Identity graphs take the guesswork out of marketing—a targeted campaign virtually always performs much better than an untargeted one. Rather than targeting everyone in a household with an advertisement that should really be meant for one person, using identity graphs can help you focus on the individual you really want to target. Advertisers can now cut down on the wasted viewable ad impressions at the top of the funnel, where they are waiting to see who turns into a first-touch engager, moving them now more precisely down the funnel and targeting them in a more relevant and contextual way.
Sometimes, a targeted tv ad might be aimed at the whole family—somewhere household data might be beneficial. Or a targeted tv ad might be aimed at a single individual watching Hulu from their personal account in their room. Either way, it helps to know the identities of the individuals who are using the devices in question—whether it’s the family TV in the living room or their personal laptop. By following the browsing activity, login information, and watch history for a certain individual, an identity graph can make it easy to guess who’s watching Netflix in the family room and who’s scrolling through YouTube videos on the smart fridge in the kitchen.
Knowing who their audience is on the individual level helps marketers avoid wasted ad spend at the top of the funnel, where they are simply creating awareness of their product. Having strong identity data at the launch of a campaign gives advertisers a chance to now enter the funnel at a lower stage, and shift some of those “awareness” funds to the “consideration” and eventual/hopeful “acquisition” phases. Rather than focusing ad spend at the top of the funnel where viewable ad impressions are most likely to be wasted, entering lower in the funnel lessens the random chance approach and allows marketers to more confidently target consumers with personalized messages and offers.
Simply put, it’s the difference between knowing that someone in a household searched for camping equipment and then serving that whole household network an ad for your tents, and identifying a device that’s much more likely to be tied to one specific person using it at that time and sending them that same tent ad. But also knowing things like age, income, and other interests for that person that is tethered to that device is critically important in being effective with your messaging.
The first phase in the rise of interconnected online ecosystems at home is just about cemented. Now, the analysis of how all these devices and individuals interact with each other, and what they are individually searching for and consuming across shared devices in their household network, will be where the market turns its attention. Through the use of deterministic and probabilistic matching, identity graphs can track how and when individuals use devices within the home, and what their personal interests may be.
Personal and individualized ads virtually always perform better than those targeted at a wide audience. Being able to inform the ads delivered to a selected household more precisely at the individual device level, which can reduce wasted impressions, acquisition time, and boost ROI, is what advertisers need to strive toward in 2022—and it can most definitely be accomplished, through the use of identity graphs. Now, the fifty-something dad and the college kid can each be served ads that relate to their interests—even when they’re in the same house, on the same network, at the same time.
At Stirista, Identity is our Identity, and we have a simple mission: help marketers generate revenue with our identity-level data. As marketers ourselves, we know that one-size-fits-all solutions and decayed data don’t work, so we built our modular and real-time OMNA Identity Graph from the ground up to arm you with marketing data that actually works.