If you only take one thing from this blog, let it be this: match rates are the beginning of the story, not the end goal. A high match rate doesn’t matter if performance is low. Accuracy and relevance of the match matters, using high-quality data to generate the match matters, having a data strategy in place and using the match as a step to enable it matters.

Let’s dive a bit deeper.

Match rates measure how likely marketers are to reach a consumer with a given vendor, most often an identity provider, but can also refer to data providers, direct with media owners or buying platforms, and any other second-party use cases. A match rate represents the percentage of the individual or household records from Company A matched to the customer records of Company B. 

Although they are frequently used as a north star metric to determine the potential reach of campaigns or match quality, a match rate percentage alone lacks context about the process. Furthermore, it is not the only indicator of the potential performance of a data partnership. Savvy marketers should enquire about how the match is calculated, especially when evaluating different providers so that they can compare like for like. To determine correlation, they should compare actual campaign results to the initial match rate and match methodology. 

What’s the issue with match rates?

Match rates are a lot like icebergs. The match rate is visible above water, but the actual process and methodology that generated the match hide underneath. You can’t successfully navigate your way around multiple match rate providers without understanding how each of their processes works.

Traditionally, creating the match requires the brand and media owner’s first-party data to be shared and commingled in a third-party location. Here the vendor can access, manipulate, and configure the match against its identity spine without giving marketers visibility or control over the process. This centralized approach requires all participating parties to sacrifice control, transparency, and often privacy of their valuable first-party data. 

Each match test can take several days to be reported back to the client as a standalone percentage. Without some key contextual variables, such as the type of data used, precision level, household composition, lookback window, type of match, or audience expansion used, marketers can have a hard time evaluating the quality and usefulness of the match.

By reporting on matched IDs that might not be addressable in the bidstream, such as internal identifiers, match rates can be easily and unintentionally inflated. Unfortunately, this is a common practice as there are often two or three match rates that need to be calculated: 

• the offline match with the identity provider’s offline graph;
• the match to their digital graph; and 
• the actual match with the media owner or bidding platform dataset. 

This final match calculation is the most valuable as it determines true reach and scale but is often not exposed in the initial match test. 

Because of this, high match rates aren’t always equal to high reach or high potential. Discarding opportunities based on the partial picture that match rates paint is not good for business, and it prevents organizations from achieving their goals, not to mention the loss of time and money. 

At InfoSum, we understand the value that ID vendors and match rates provide to marketers. However, we believe there is a better way to leverage that data and that match rates alone don’t paint a complete picture of the true addressability of a campaign, especially as unstable IDs like cookies and device IDs that account for the false scale are soon to be gone. We want to empower marketers to make match rates truly work for them by providing complete transparency so they can avoid the icebergs. Our goal is to help marketers ask the right questions to understand and extract the full value of their data and evaluate how to achieve the best performance.

To evaluate any match rate, marketers need to understand: 

1. Their strategic goal and the role the match rate plays vs. other metrics, such as reach
2. The quality and composition of the datasets being used to match
3. The precision level (individual/household etc.) and the freshness of the data
4. The difference between match tests and actual campaign performance like ROI, ROAS, CPC, and CPA
5. When to apply deterministic or probabilistic audience expansion

Reframing match rates for the new age

To further complicate things, the advertising industry is currently in flux, with the disappearance of third-party identifiers and the emergence of new media channels such as retail media, connected TV, audio and more. Accurate and actionable match rates with an identity provider are already suffering in markets such as the EU due to privacy legislation. Globally, match rates are being impacted by technology changes and a lack of high-quality consented data. Clear and warmer waters are ahead - direct match rates between the buy and sell sides are becoming increasingly important to enable data-driven strategies. 

Luckily, there are no icebergs in these waters. Direct data matching between a brand and a media owner empowers marketers to understand their audience and tailor advertising in a relevant and considered way. As it is one-to-one matching against a known audience and without a daisy chain, ad waste is minimized by improving the accuracy of target profiles and maximizing the chance of achieving business outcomes. Similarly, media owners can maximize their yield with higher CPMs and greater control over their first-party data. 

The benefits of leveraging first-party data by both the buy and sell side in a privacy-safe way include:

• Providing a sustainable currency for digital advertising even when third-party identifiers disappear entirely.
• The ability to power highly accurate and dependable targeting. 
• Allowing marketers and media owners to truly understand and address their audiences.

Making match rates work for you

With the match rates iceberg melting, brands and media owners can have more visibility under the water and prevent hitting any unwanted outcomes. Based on our experience, the keys to making match rates work for your goals are:

• Having a clear strategy and business goal in mind 
• Working directly with activation partners when possible
• Focusing on using high-quality data vs. high match rates
• Using a secure collaboration technology to enable the match (like the InfoSum data clean room)
• Ensuring you know exactly how the match is enabled
• Creating actionable matched segments
• Increasing reach using audience expansion tools and leveraging ID vendors strategically

At InfoSum, we've seen the power of direct data matching, with clients achieving outcomes equal to and often exceeding those possible through traditional methods, regardless of the original match rate. Explore our case studies to see how first-party data strategies can deliver the right performance.