Why We Need to Change the Conversation Around Match Rates in Digital Advertising

Match rates have long been important for marketers. After all, no matter the vertical you operate in or the channel on which you plan to advertise, reaching the right audience is critical to achieving your goals. However, the digital advertising industry is fast evolving, and the way we talk about match rates – and how exactly we rely on them – needs to change.

Below, answers to some key questions about the role of match rates and identity resolution for today’s marketers.

What Exactly is a “Match Rate?”

Match rates often refer to the percentage of users from an audience segment that a demand-side platform (DSP) is able to recognize. Match rates have long been important, because they allow advertisers to understand the size of their addressable audiences – in theory, when matching apples to apples, the higher the match rate of a specific audience segment, the larger the number of consumers who can be reached with digital advertising – in other words, the more DSP addressability for that audience segment.

How are Match Rates Calculated?

The formula for calculating match rates used to be relatively simple: cookies and/or devices were matched between two parties either cookie to cookie or device to device. The results were clear; a higher match rate meant more cookies or devices, which meant higher external addressability for DSPs.

However, moving to true people-based marketing has introduced a slate of new variables beyond cookies and devices. Now, marketers are able to leverage location data, IP addresses, email addresses and more, which require the identity resolution process to power addressability for media across devices. Thus, a match rate is no longer a simple formula providing that easy, clear comparison metric the industry has been reliant on since the invention of the DSP.

What is Identity Resolution?

Identity resolution is the process of resolving data back to an individual or household by matching personal identifiers (for example, email address or name) to digital identifiers (like cookies or IP addresses) across devices and channels. Because of these new data points that are necessary for addressability in people-based marketing, diligent privacy compliance is mandatory. In addition, because reaching target audiences across all channels including linear and connected TV, audio and digital-out-of-home (DOOH) is essential, the identity resolution process is an integral need for any DSP today. In fact, omnichannel DSPs are best equipped to solve for identity; as we discuss here, working in a unified omnichannel DSP helps advertisers have a better understanding of just who they have the ability to reach.

True identity resolution is complex, however. It requires the ability to distinguish relevant touchpoints from noise or waste, and to resolve them into organized, anonymous profiles, which are then anchored in a comprehensive device graph.

Hear Industry Thought Leaders’ Predictions for How Identity Resolution will Shape Marketers Plans in 2021. (See survey results)

Are Match Rates “Moot?” Is There a More Important Metric?

The simple match rate comparison metric is becoming a moot point only relevant with the legacy cookie sync.

The issue with using a match rate metric to compare DSP platforms in a people-based ecosystem is that it is no longer that simple formula referenced earlier of comparing apples to apples. The reality of a DSP match rate is that it is dependent on the specific identifiers or data points being matched between both parties at the specific moment that matching occurs.

Data points fluctuate at varying levels between parties based on how they are uniquely collecting data, number of opt-ins and more. This ultimately dictates how many data points are currently available at that moment. In simpler terms, it is no longer matching apples to apples – or in industry terms, cookies to cookies – and thereby no longer a credible metric for effectively comparing DSPs.

Match rate formulas will differ between platforms as each will take its own approach based on the type of data it manages, how it manages that data, favored logic or perhaps what calculation shines the best light. Additionally, after the match occurs, identity resolution is needed to map all the disparate data points that were matched back to a respective anonymous profile, which powers addressability.

Is Identity Resolution a Good Means to Compare DSPs?

Yes. People-based marketing, which involves many varying types of data points from a multitude of sources, is the most effective approach for an audience-based approach to media. Identity resolution is necessary for people-based marketing and is one of the most critical aspects powering media addressability.

Some DSPs – like Viant – offer full identity resolution capability along with the ability to activate media. Others depend on external identity resolution providers. That’s one key question to ask any potential DSP partner.

What Were Match Rates Really Solving for Marketers?

Ultimately, match rates initially helped marketers understand how much scale a DSP could provide for their target audience segment. This is really the focus to keep discussions around. It’s not about the match rate for the sake of a match rate – it’s about how much of the target audience segment a DSP will be able to reach.

When evaluating DSPs, advertisers need to be focusing on what we’ll call “matched universe output,” that is: how many addressable profiles (either households or individuals) are matched and reachable for omnichannel media activation. With the death of cookies on the horizon, it is important that advertisers adapt their evaluation considerations and understanding in order to effectively answer the same important questions in a new, more complex reality.