Defining zero-party data 

Editor’s note: Steven Millman is SVP global research and operations, advertising solutions, at data and insights platform Dynata. 

Zeroth-party data is a relatively new term in the market research industry which appears to have originated at Forrester in 2018 (account required to access) and describes a way to divide what has traditionally been referred to as first-party data into two groups based on the conditions under which the data were collected. 

The name “zero-party data” is inconsistent with the nomenclature. Given the use of ordinal terms for the other types (first, second, third), the only really appropriate name for this would be zeroth party data. Using the cardinal zero fails to align with the rest of the existing nomenclature and confounds one’s sense of mathematical consistency. I do math for a living, so I’ll be using “zeroth-party data.”

Before I dive further into zeroth-party data, let’s take a brief look at first-, second- and third-party data. 

First-party data refers to data that a brand or research firm collects directly, owns and has implicit or explicit rights to use. Implicit rights might include those that arise from a brand’s direct relationship with its customers, like sales or other CRM data, e-mail lists or website interactions. Explicit rights are those granted voluntarily by the source of the data such as survey data, voluntary meters or accepted cookies used by the publisher (as opposed to some third-party entity).

Second-party data is not collected directly by the user of the data but rather is obtained by them from the legitimate first-party owner. This can be as simple as buying the data from a first-party owner who has the rights to sell them or can be the combination of two first-party data sets where both owners have the rights to combine them. When obtaining second-party data, it’s in your interest to be very cautious and verify that the data owner or owners do in fact have these rights.

Third-party data is collected, aggregated or purchased by entities that do not have any direct relationship to the legitimate owners of the data. Examples of third-party data include accepted cookies served on a website by parties other than the publisher, syndicated survey data, purchased meter data and most commercially available location data.

Zeroth-party data: A subset of first-party data

With all that in mind, on to zeroth-party data. Zeroth-party data is a subset of first-party data in which the data collected come from asking direct questions of individuals about their preferences or opinions rather than related to behaviors and personal facts, and never through passive observation of their activities. 

Zeroth-party data also presumes the existence of a value exchange – that the consumer receives some form of benefit from sharing their beliefs. Forrester’s definition of zeroth-party data also refers to these data as being specifically related to brands and not research companies (more on that later). 

“Zero-party data is that which a customer intentionally and proactively shares with a brand. It can include preference center data, purchase intentions, personal context and how the individual wants the brand to recognize her." Forrester Report

In this context, telling a panel your age, gender and mailing address is first-party data whereas your opinions and preferences about a brand are zeroth-party data. Observed data for which one has explicit permission to use (e.g., meters or first-party cookies) would also be first-party data. 

The Forrester article describes zeroth-party data as self-reported data provided directly to or on behalf of a brand which should never be shared or sold afterwards. The Forrester authors suggest that customers “feel deceived if a brand sells or shares their sensitive zero-party data. When a consumer shares zero-party data with you, she's expressing trust and a willingness to be vulnerable to your brand – but not to a third party that might be purchasing the information from you.”

Credible survey panels do not ever use panelists or their data for direct marketing purposes and as a result, the vast majority of the data we collect by survey would fall into the zeroth-party data category. Although survey firms sell self-reported data to their clients, in most cases clients are paying them to collect it on behalf for their exclusive use. Syndicated surveys on the other hand, which are resold over and over again to multiple clients, would be first-party data.

Therefore, zeroth-party data can be generally defined as:

• A subset of first-party data.
• Exclusively derived from voluntary self-report.
• Removeable at the request of the contributor.
• Part of a value exchange between consumer and brand.
• Explicitly or implicitly not for resale to third parties.

Privacy-centric data 

The purpose of creating a term like zeroth-party data appears to be as a warning to brands to be cautious about reselling their interactions with customers to other parties, especially when those are opinions or beliefs. These sorts of interactions also provide marketers with cookie-less, privacy-centric data with which to improve targeting and ad relevance. It is generally a good privacy policy to not resell consumer data from self-report to marketers, with or without the new term, and certainly safer from a legal standpoint as the pendulum continues to swing – appropriately – toward the protection of consumer privacy.

Finally, be cautious about commercial marketing materials making use of this new identification. Zeroth-party data of late has started to be misused in the marketplace as a catchphrase without a clear understanding of the term’s original meaning and purpose. This has been particularly noticeable among a subset of data collection companies.