Embrace complete customer views you can act on — with identity resolution and CDP

Are you unknowingly overlooking your own high-value customers? As it turns out, most brands are.

Imagine you’re a retailer, and a customer has purchased your product in-store, then browsed your website as an anonymous visitor. They later engage with you on social media using an account associated with an unknown email address. This customer is engaged across all of your marketing forums and has the potential to be a high-value consumer. If your POS data, social media analytics, and web data are siloed, you end up with fragmented  identity an incomplete customer view, and a missed opportunity to engage with a consumer that would be likely to purchase again. 

Imagine that digital impression on social media led to another offline conversion. In the current fragmented customer view, you would have no way of associating those insights with a consumer. When leveraging identity resolution, you are able to tie all of your disparate sources of attributes to individuals for marketing, retargeting, and analytics. Enter: identity resolution (and CDP technology).

The biggest mistake brands are making with identity management

2022 data from Experian shows that 84% of businesses today agree that recognizing their customers across channels and over time is extremely important. 

• But as many as 30% of retailers admit they cannot accurately track consumer behavior across devices.
• 71% of marketers have difficulty maintaining an accurate customer profile over time.
• And 84% of consumers agree retailers should do more to better integrate their online and offline channels

So, what’s the problem, exactly? Kevin Dunn, VP of Industry Sales, Retail & CPG, at LiveRamp describes it well. 

If you have multiple identity frameworks, you essentially have no framework (and you also have very little meaningful insight).

“We see more and more  brands making the mistake of having a fragmented identity framework: typically one in the online or unknown space, one in the offline or known space, and three or four different applications or data infrastructures. We see it often amongst brands that believe one application or platform encompasses all the functionality they need – such as CDPs, DSPs, ESPs, Clouds, etc.,  but quickly realize they aren’t unifying the known and unknown into one solidified view for mitigating media waste,” says Dunn.

“Most brands don’t go all-in with one best-of-suite technology stack. They want the best they can get in every category. But for that, you need a powerful identity framework to give you a unified customer view across that ‘best-of-breed’ stack.” 

The issue with treating this fragmented data as the business and customer insight you act on is that you then tend to have repeat messaging, oversaturation of messaging to customers, or risk turning customers off entirely because their experience isn’t consistent. 

Customers don’t think or behave in channel silos, so why should we? At LiveRamp, we talk to our clients about putting an identity foundation at the core of the creation of their customer profile. But the challenge many of them face is: how to channel seemingly disparate data into the right identities to ensure the most accurate, consolidated identity information in one record. 

Most marketers struggle to get this right, which unfortunately means every campaign they launch runs the risk of being built on and targeted, leveraging  poor quality or inaccurate data. According to MarketingProfs, even your most loyal customers likely have at least six different identities within your marketing system — and some audience members may have over 100. When you have anonymous users that are engaging with you without registering or establishing a login, the challenges are even more complex.

Four ways data-centric companies can master identity resolution

Identity resolution plays an important role when it comes to your organization’s data-driven goals. It’s the key to truly understanding your customers.

Modern businesses need a solution that’s prescriptive, one that allows a brand to reach a customer with confidence — rather than reaching them across four devices believing they’re four different individuals , simply because that single profile isn’t properly stitched together.  Here’s how to prepare your organization to get it right.

1. Understand deterministic vs. probabilistic matching

Understanding the difference between deterministic and probabilistic matching helps determine your approach to identity matching.

Deterministic matching resolves identities based on what you know about the customers, using data such as phone numbers, user IDs, email addresses or device IDs to match and merge records across multiple touchpoints. It reflects a high level of confidence in the data you have and the decisions to merge records, resolving multiple data sources into one record.  LiveRamp’s graph is a fully deterministic graph in both the online and offline infrastructures.

Probabilistic matching resolves identities based on predictions of what you infer to be true. Using predictive algorithms, probabilistic matching gives you a sense of likelihood that your customers are the same in disparate, multiple records. Most probabilistic matching statistical models also provide you with a confidence interval, which indicates how probable it is that records belong to the same user.

2. Don’t underestimate the role of the identity graph

Identity resolution begins by bringing all of your disparate sources together while leveraging a deterministic identity graph as a source of truth.  You will allow for both known and pseudonymous databases to power various use cases of both targeting and analytics ensuring you understand who your customers are both at the individual and household levels. This will allow brands and marketers to have their own version of their personal identity graphs.

• There may be thousands, millions, or billions of identifiers in the identity graph. 
• These pieces of information – fragments – may have some interrelated connections, meaning they contain multiple pieces of data. 
• These fragments may or may not be tied to a customer profile.

In real time, the best platforms for identity resolution will connect a piece of information from one source to another. For example, an email address used to register for a newsletter may be the first information source used. The information graph may then discover that the same email address was used to complete a purchase on your organization’s website. That single identifier, the email address, can then be used to enrich the profile already in your system for the transaction. At LiveRamp, we underscore the importance of people-based identifiers, and help make cross-channel profile creation  both possible and accurate, with RampID.

Identity graphs are the bridge between offline and offline experiences, so having an offline foundation is critical. However, you also need the ability to layer on your own logic, as well as the ability to act on the insights. Because without action, what good is a database full of profiles? That’s where, alongside your strategy for identity resolution, a CDP like Lytics comes in.

Lytics CDP is built on an identity graph that enables flexible and intelligent stitching. When user behavior across devices is consistent, the graph associates that user’s activity together to deliver experiences based on all of it. For example, if a user starts reading a New York Times article at home on their tablet, continues on their  phone on the  way to work, and then finishes on their office laptop, that behavior is connected within the graph. And when behavior across devices isn’t consistent, Lytics is set up to recognize and resolve the inconsistencies as well. 

3. Act on your complete profiles with identity resolution and CDP

There’s a beautiful synergy to be found in CDPs and identity graphs. A CDP helps you build an identity logic with your own first-party data. and then an identity graph layers on external knowledge of who that person is, what touchpoints they use, etc. Ultimately, you end up with a de-siloed collection of first-party data and a central place to collaborate with other data sets ultimately making your data more intelligent. 

Once you’ve achieved this, you may want to think about acting on this insight by, for example, suppressing audiences unlikely to engage. 

To power the most accurate and advanced suppression use cases, an identity graph is critical in helping you determine who to suppress in advanced targeting initiatives., Unified identities provide a fuller, more comprehensive profile of the customer: allowing marketing teams to reach people in more markets, more product lines, and with more specific information that’s both resonant and relevant.

4. Layer third-party knowledge onto your first-party insights, with data clean rooms

Second-party data is a hot topic among marketers, and often misunderstood. By definition,  it’s simply the data you get from the merging of two first-party datasets. Finding a way to  overlap these two distinctly sensitive and unique data sets, securely, for a long time was a major challenge for businesses. Today, we see a growing adoption of data clean rooms, which  provide a secure way for brands and their partners to combine first-party data, share it across clouds or clean rooms, and use it to create interesting insights for targeting,monetization, retargeting, and more. 

Lately, we most frequently see CPG brands and retailers working together on these first-party data sets using clean rooms, in order to power co-branded targeting initiatives for increased category adoption of a product, like remarketing to consumers who have purchased the product before   as well as product awareness advertising for potential customers who haven’t. This approach allows brands to monetize their first-party data, one of the most powerful forms of consumer data, by layering on more insight and context without compromising security.

How Lytics brings order to identity chaos

Lytics is designed to bring order to the chaos that can come from harnessing known and unknown identities. Lytics simplifies identity resolution by creating profiles backed by identity graphs and profile stitching. This means if you have a Lytics tag that lives on your site, that’s going to drop a first-party cookie on a person — even if that user jumps to a different device, you can then re-associate that person. This is true even if you don’t know who they are.

Lytics captures a variety of consent-based identifiers — collecting and merging user identities based on deterministic and other matching techniques. Using a dynamic merging approach, Lytics maintains data integrity, giving your team the flexibility to test and revise your personalized marketing strategies across your entire marketing tech stack. 

• Unlock the power of stitching cross-channel data fragments into cohesive user profiles, while uncovering important behavioral data and topic affinities for each user. 
• Enable large volumes of robust profiles, which can be resolved as new events come into Lytics in real-time
• Ensure up to 99.7% user profile accuracy by matching and merging identities across web, email, and social media
• Guarantee meaningfully personalized experiences with rich, complete user profiles.

Want to dive even deeper? Explore Lytics technical documentation to learn the ins and outs of how the CDP manages identities.

With complete and accurate customer views powered by identity resolution, you can leverage your CDP to identify your most engaged customers with a few clicks and set up a marketing experience just for them with a few more.

And, importantly, with Lytics, all of this happens in real time. You’re never basing personalization on week-old information. Lytics is built on an identity graph that is constantly importing and stitching together new user behaviors and information as they happen, automatically and at scale.

Lytics and LiveRamp: Better together

What is the value of taking a CDP dealing predominantly in first-party data and pairing it with an identity solution like LiveRamp to create a complete identity framework for an organization? Data collaboration and unification.

When you take your understanding of your customer’s behavior and layer on additional demographic, psychographic and touchpoints you take your understanding of that customer to the most advanced level possible for the best personalization experience.  For example, if you’re a CPG brand selling household products to families, you may want to stitch together individual or household level information. Pairing an identity graph with Lytics CDP (including capabilities for tag management, site personalization, in-app functionality, and audience segmentation) allows marketers to not only target and personalize to that household, but understand all of the individual profiles who reside together.  LiveRamp, a powerful platform designed to enable data collaboration across organizations in a privacy-first, interoperable, neutral way. For an all-in-one, powerful and integrated solution, your organization can pair Lytics CDP’s behavioral data and content affinities with LiveRamp’s identity resolution technology for meaningful personalization and engagement that converts prospects into loyal customers.

For even more on how Lytics and Liveramp make identity management mastery possible, watch the on-demand webinar recording in full.