Zero-Party Data

Zero-party data is information a customer chooses to hand you on purpose. Examples: a quiz that asks about skin type, a preference center where someone picks email topics, a shoe size entered on a profile, a post-purchase survey that asks why they bought, or a form field asking when a project needs to start. The term was coined by a research firm to separate it from first-party data that you merely observe, like pages viewed or items clicked. The person knowingly said it, and usually understood why you were asking. Good questions are short, tied to a benefit, and specific enough to act on, such as budget range, timeline, skin concern, or skill level, instead of open-ended essays.

The mechanism is a fair exchange. You offer something useful in return: better recommendations, a discount, a tailored guide, a faster quote. The answers are then stored against an email address or account in your CRM, email platform, or store. From there they can drive segments and ad audiences. Upload a list of customers who said they have oily skin as a Customer Match audience in Google Ads or a custom audience in Meta, and the creative can speak to that exact need. Nothing about the data is inferred, so the targeting is more precise than a model's guess. Tools such as quiz builders and form tools can write answers into profile fields or tags in your email platform, which then sync to ad audiences through list uploads or integrations.

Technically it is a subset of first-party data, since you collect it directly. It follows the same rules for ad use: you need to disclose how the answers are used, and in regions with consent rules, uploading them to an ad platform may need consent. Ad platforms do not treat it as a separate class, and it carries no special legal status. What changes is how you plan around it. It works best for sorting audiences, choosing which of several offers to show, and writing copy, and it works poorly as a source of scale, because only a small share of visitors will complete a quiz. Because answers arrive with a clear purpose, you can honor it by tailoring what follows. Ask someone's goal and then send content about that goal, or the exchange feels like a trick.

Mistakes follow a pattern. Asking too much on the first screen drives people away, so start with one or two questions. Collecting answers and never using them wastes the exchange and annoys people when the follow-up ignores what they said. Trusting stated preferences blindly is another trap, since people say they want one thing and buy another, and answers go stale. Finally, treating the term as proof of consent is wrong: a person answering a quiz has not agreed to be advertised to unless you asked and told them. Also treat self-reported data as a signal rather than a fact. Someone who says they plan to buy next month may not, and a household's stated budget can be aspirational. Combine it with behavior.

Check it by comparing segments. Take the group that answered a quiz question and the group that did not, and compare conversion rate and revenue per customer in your email or store reports. This week, add one question to an existing touchpoint, such as the order confirmation page or the first welcome email (for example, what are you shopping for?), store the answer as a tag, and build your first segment from it. Then write two ad variations, one per answer, and see which segment responds. Refresh the questions yearly and let people update their preferences, because old answers decay.