Since their advent in the ‘90s, cookies have been considered a problematic solution to the complicated problem of gathering data on user-level behaviors. Cookies are problematic because they aren’t directly connected to people, they don’t accurately capture frequency, and they flat out fail at recognizing cross-device usage, not to mention that they invade user privacy. As cookies go away, and we choose what to do next, should we replace one problematic approach with another that mimics it and potentially inherits all the same challenges?
When it comes to building data-sets for analytics, there are two main options: some sort of cookie replacement, or using permission-based panels. The first article in the series laid out the key aspects of the two options, and our second article explored the challenges inherent in working with the big data approach cookie replacements rely on. In this article, we’ll focus on why panel-based approaches are faster, more accurate, and more affordable.
Panels have a history of being too small to be useful, with questionable representation and weighting schemes that dramatically skew insights based on very small changes to the sample. Large, representative panels used to be cost prohibitive, but not anymore. Modern panel-based approaches have moved beyond this.
With a small super-computer now in everyone’s pocket, with numerous apps powered by sensors running 24/7, we have access to many more potential panelists, with much more robust data collection mechanisms that provide passive tracking with more accuracy and at a lower cost. Because these panels are built with the explicit permission and participation of the users, we are able to gather greater insight not only into passively tracked behaviors, but also into the beliefs that underpin these behaviors. Many of these panels have constant, active recruitment in place to ensure even small niche audiences can be accurately measured on a continuous basis.
As part of the processes to recruit a representative panel and gather data on beliefs and behaviors, panel-based data companies have also built-in quality checks and normalization processes. These quality checks filter out bots and survey farms, and account for extreme outliers. The result is insight from a panel whose data aligns with real world facts and observations. This clean, consistent data set makes it much easier for marketers to connect with external data sources, like third-party audience enrichment or large publishers.
Standardized approach allows linkage with external sources like Google
One of the benefits of panel-based approaches is that they are more standardized. This creates stability for linking with external partners including walled gardens (like Google and Facebook), ad networks, and other media outlets. Kantar’s Project Moonshot has created stable, always on, direct integrations with the most popular publishers (including Google/YouTube) for real-time exposure data. This type of seamless integration isn’t possible with solutions built on non-permissioned users, because the largest publisher platforms are not willing to expose themselves to the privacy liability of tracking their users and sharing that data without their explicit permission.
The Kantar is home to the world’s largest audience network. Through our extensive profiling efforts and data connections, we have more information about the people marketers want to reach. And in a world where the value of first-party data is rising as the reliance on third-party cookies diminishes, we deliver data from online research panels you can act on, quickly.
Our API-driven ecosystem of more than 150 million compliant panelists spans 130 global markets and includes consumers, B2B audiences, healthcare professionals, patients and non-survey-based data connections. You can access this repository of knowledge quickly and with ease for precise targeting and enhanced insights. By pairing our vast network of compliant market research panelists and data sources with tech-enabled research solutions, Kantar delivers unrivaled access to quality panel and first-party data with speed and at scale.
A quality approach that is vetted
Kantar’s extensive expertise in building representative panels has delivered a dataset with up to 4,800 rich profiling attributes per panelist. Our investment in anti-fraud measures ensures that these are real people with real attributes. We authenticate respondents through IP address, email, and physical addresses.
We also use AI and machine learning to flag people who are predicted to be poorer responders or potentially dishonest respondents before they join the panel. Out of this uber-panel we have consumer panels, B2B panels, healthcare and more. All of these capabilities are available in dozens of countries and languages, creating a turn-key solution for global brands. Kantar’s panel build-out capability is so strong that the Association of National Advertisers (ANA) chose Kantar to pilot test alternatives to Nielsen in the US, which isn’t surprising given Kantar currently conducts TV measurement in 61 countries.
Marketers should be consumer centric
At the core of the marketer’s job is an obsession with the customer. Understanding customer needs and pain points leads to insight for product innovation and messaging strategies that grow businesses. This same empathy must also be applied to marketing’s approach to gathering data about consumers. Data-collection schemes like cookies, that rely on consumer ignorance, laziness, or even downright deception have no place in a true customer-centric toolbox. As the industry adapts to a world without cookies, the winning solutions will be those that put privacy first and use that permission to gather rich data on consumers, their context, preferences, beliefs and behaviors. Analytics built on these data sets will be easier to work with, less expensive, and ultimately provide more accurate recommendations. For all these reasons, Kantar believes permission-based panels are the foundation of a promising future for marketing analytics.