Definition and examples
The test is the relationship. Data is first-party when the business gathered it directly from its own customers, users or visitors, through its own channels. Typical examples:
- Transactions: orders, products, values, returns, subscription status.
- Accounts and sign-ups: email addresses, phone numbers, preferences given at registration.
- Engagement: email opens and clicks, app usage, loyalty activity.
- Site and app behaviour: pages viewed, searches, items added to a basket, recorded by the business's own analytics.
- Sales and service records: CRM stages, support tickets, survey answers.
Where it is stored does not change its status. Customer data held in a CRM, a data warehouse, an email tool or a customer data platform is first-party in each case, because the business collected it.
First-, second- and third-party data
The three labels describe the route the data took. Second-party data is another organisation's first-party data, shared with the business under a direct agreement, such as a partner sharing an audience. Third-party data is assembled by companies with no direct relationship to the people described, from sources such as cookies, device identifiers and partner feeds, and sold or licensed to many buyers. The comparison of first-party and third-party data sets out when each fits and why the balance has shifted towards data the business controls.
In practice: audiences, measurement and personalisation
First-party data does three jobs in marketing. It builds audiences: customer lists uploaded to ad platforms to reach existing customers, to exclude them from acquisition campaigns, or to seed lookalike audiences. In programmatic buying, publishers' own reader data is a reason to buy directly, as the programmatic versus direct buying guide notes, and advertisers' own lists are the basis of remarketing and suppression.
It improves measurement: hashed customer details sent with a conversion let ad platforms match conversions that browser identifiers lost, and CRM outcomes sent back to the platforms tell bidding which leads became customers. The Google Ads measurement hub and the guide to connecting a CRM to ads and analytics cover those routes. And it drives personalisation: recommendations, lifecycle emails and on-site content based on what the customer has done.
Where it lives is a design decision. The marketing stack design guide describes the current pattern as first-party data in a warehouse, a customer data platform or both, replacing stacks built on third-party cookies.
Consent, quality and reach limits
Owning the data does not settle what may be done with it. Personal data collected directly is still subject to the consent and lawful basis the business holds and to what its privacy notice told people; the privacy, consent and measurement guide covers the obligations in general terms. Data collected without recording consent is hard to use later, however valuable it looks.
Quality is the second limit. Inconsistent forms, missing fields and duplicate records reduce how much of the data can be matched or analysed. Reach is the third: first-party data describes people who already know the business. It is the best material for retention, measurement and suppression, and on its own it cannot find new customers who have never visited.
A customer list that matched less than expected
Marlowe Pet Pharmacy had 18,000 repeat-prescription customers and wanted to stop showing acquisition ads to them. It uploaded the list to its ad platform as a customer-list audience, but only about 7,200 records matched, around 40%. A review found that the prescription form treated email as optional, so 6,500 records had none, and older records held work addresses customers no longer used. The pharmacy made email a required field with a clear consent statement on the form, and asked existing customers to confirm their details when they next ordered. Eight months later the list held 21,000 customers, 16,800 with consented email addresses, and 10,900 matched. Excluding them from acquisition campaigns reduced spend on existing customers by about £900 a month.
Related terms
A CDPUnified first-party customer profiles. is one common place to unify first-party data from many systems into customer profiles. A Custom AudienceAn ad audience built from the advertiser's own sources: customer lists, site visitors or engagers. built from a customer list is the most direct advertising use of it. Enhanced ConversionsA Google Ads feature that sends hashed first-party customer data to improve conversion matching. sends hashed first-party details with a conversion to improve matching. The restriction of Third-Party CookiesCookies set by a domain other than the site visited, used to recognise a browser across sites. by browsers is the main reason first-party data has moved to the centre of targeting and measurement plans.
Common confusions
- First-party means stored in our systems. It means collected directly by the business; bought data in the warehouse is still third-party.
- Owned data needs no consent. Collection and use still depend on consent and lawful basis.
- First-party data replaces prospecting. It reaches people who already interact with the business.
- More data is better. Fields nobody uses add risk without value.
- Upload rates equal match rates. Platforms match only records they can link to their own users.
