Zero Party Data That Customers Will Actually Give You
Zero party data is what a customer tells you on purpose: sizes, skin type, the person they are shopping for, how often they want to hear from you. It is the most accurate data a store can hold and the easiest to ruin with a greedy form. Every field you ask for costs completion, so the asks have to earn their place by visibly changing what the customer sees.
The exchange has to be visible
Customers answer questions when the payoff is immediate and obvious. A two question fit finder that narrows 200 products to 12 earns its answers. A nine field quiz that ends in an email gate and a generic bestseller list does not, and completion data shows people abandoning it in the middle questions where the effort stopped feeling reciprocal.
Ask at the moment the answer changes something. Size and fit questions belong on category and product pages, gift recipient questions belong in a gift flow, replenishment timing belongs after a purchase, when the customer actually knows how fast they use the product.
What to collect, in priority order
Start with the attributes that drive your assortment decisions: the size system, the key fit or skin or dietary constraint that splits your catalogue in half, and the use case. Those three answer types power recommendations, filters, and segmented email without needing anything sensitive.
Identity adjacent data needs a harder justification. Birthdays are collected by everyone and used by almost nobody beyond a discount email the customer did not ask for. If a field does not change a recommendation, a reminder, or a price, delete the field and watch completion rise.
Storage is where most programs quietly fail
Answers dumped into a free text tag or a quiz tool that does not sync to the customer profile might as well not exist. Zero party data belongs on the profile as structured attributes with the same names your recommendation logic and email segments use, set once and reused everywhere.
Date it, too. Preferences decay. A size recorded two years ago, a gift recipient from last Christmas, a shade matched to a discontinued line: stale zero party data produces recommendations that feel creepier than no personalization at all, because the store clearly remembered and clearly got it wrong.
Refresh without nagging
Build the refresh into natural moments: a one tap confirm or change prompt on the account page, a pre filled preference check inside a replenishment reminder, a post purchase question that asks one thing while the customer still cares.
One question at the right moment outperforms a yearly preference survey almost nobody completes. And every answer a customer updates themselves is a data quality fix you did not have to guess at with a model.
Keep reading
Related guides on this site: Personalizing for Anonymous Visitors: What a First Session Can Honestly Tell You Frequency Capping Personalized Offers: When the Third Reminder Stops Being Personal Practical answers about personalization.
Frequently asked questions
What is zero party data?
Information a customer intentionally shares with a store, such as sizes, preferences, and use cases, usually through quizzes, preference centres, and post purchase questions. It is distinct from behavioural data a store observes or infers.
How many questions should a product quiz ask?
As few as will change the recommendation, commonly two to five. Completion falls as quizzes lengthen, and every question should visibly affect the results the customer is shown.
Where should zero party data be stored?
On the customer profile as structured attributes that recommendations, search filters, and email segments all read. Answers trapped inside a quiz tool or a free text tag cannot be acted on reliably.