The 10 Types of First-Party Data (With Examples)
First-party data is any information a business collects directly from its own audience and properties, with consent. It comes in ten common types: identity, behavioral, transactional, declared, engagement, support, account, loyalty, offline, and contextual data. Most businesses already generate all ten without realizing how much they hold.
First-party data is more than an email list
When people hear first-party data, they usually picture an email list. That is one slice of one type. In practice every interaction a customer has with your business produces data you are entitled to keep: what they looked at, what they bought, what they told you, what they complained about, and how they responded. The problem is rarely that you have too little first-party data. It is that the data sits in separate tools that do not talk to each other.
This guide breaks first-party data into ten types, with concrete examples and a note on where each type usually lives so you can find it. Then it covers how the types map to outcomes, the mistakes that keep them siloed, and what it takes to turn ten disconnected streams into one usable foundation.
Quick definition: first-party data is data you collect yourself, directly, with permission. Because you collect it, you own it, you control it, and you can use it without depending on a third party that may disappear.
1. Identity data
Identity data is the connective tissue. It is the set of stable identifiers that lets you tie every other type of data back to one real person: names, email addresses, phone numbers, postal addresses, and the account or customer IDs your own systems assign.
Examples: a customer record in your ecommerce platform, a CRM contact, an account login, a hashed email captured at newsletter signup. Identity data is what makes the difference between a pile of anonymous events and a profile you can act on.
- Where it lives: CRM, ecommerce platform, auth system, email tool
- Good for: stitching everything else together, suppression, deduplication
- Watch out for: duplicates and near-duplicates that fragment one person into several records
2. Behavioral data
Behavioral data is what people do: pages viewed, products browsed, search terms entered, features used, videos watched, items added to cart, clicks, and scroll depth. It is the largest-volume type most businesses generate and the one that reveals intent and interest in close to real time.
Examples: a shopper who views the same product three times in a week, a trial user who never opens the core feature, a reader who finishes long-form guides but skips the pricing page. Each pattern is a signal you can act on while it is still fresh.
- Where it lives: first-party analytics, product analytics, your own event pipeline
- Good for: intent signals, personalization, timing, churn prediction
- Watch out for: capturing it only into ad pixels you do not control instead of systems you own
3. Transactional data
Transactional data is the record of what people actually paid for: orders, purchases, refunds, subscriptions, renewals, upgrades, downgrades, and payment history. It is the highest-signal data you own because money is the clearest expression of intent there is.
Examples: average order value, time between purchases, the product someone always reorders, the subscription that downgraded right before it canceled. Transactional data anchors lifetime value and is the backbone of any serious retention or expansion program.
- Where it lives: ecommerce platform, billing system, point of sale, ERP
- Good for: lifetime value, retention, reorder timing, high-value segments
- Watch out for: leaving it locked in the billing tool, disconnected from behavior and support
4. Declared (zero-party) data
Declared data is what customers tell you on purpose: preferences, intentions, goals, and context they volunteer through quizzes, surveys, onboarding questions, and preference centers. This is also called zero-party data, and it is uniquely valuable because it is accurate by construction. The customer chose to share it.
Examples: a skincare quiz that captures skin type, a B2B onboarding question about team size, a preference center where someone selects the topics they want. You cannot infer this data reliably from behavior, which is exactly why asking for it directly is worth the effort.
- Where it lives: survey tools, quiz tools, preference centers, onboarding flows
- Good for: personalization, segmentation, product fit, reducing guesswork
- Watch out for: asking for it without giving an obvious benefit in return
5. Engagement data
Engagement data captures how people respond to your outreach: email opens and clicks, SMS responses, push notifications tapped, ad and campaign interactions on your own properties, and webinar or event attendance. Where behavioral data is what people do on your site, engagement data is how they react to what you send.
- Where it lives: email platform, SMS tool, marketing automation, campaign tools
- Good for: deliverability hygiene, re-engagement, channel preference, sequencing
- Watch out for: treating opens as truth in a world of inflated open-rate signals; weight clicks and replies higher
6. Support and service data
Support and service data is everything that happens when a customer needs help: tickets, chat transcripts, call notes, returns, warranty claims, and complaints. It is often the richest source of churn-risk and product-gap signal you own, and it is almost always trapped in a help desk that nothing else reads.
- Where it lives: help desk, ticketing system, contact center, returns portal
- Good for: churn risk, satisfaction, product feedback, save-the-customer plays
- Watch out for: a support history that the marketing and sales teams cannot see
7. Account and profile data
Account and profile data describes how a customer has set themselves up to use you: settings, saved items, plan tier, seat count, role, integrations connected, and account configuration. In software especially, this is a strong predictor of value and stickiness.
- Where it lives: product database, account settings, admin console
- Good for: expansion signals, onboarding completeness, fit scoring
- Watch out for: configuration that signals risk (no integrations, single seat, empty profile) going unnoticed
8. Loyalty and rewards data
Loyalty and rewards data comes from programs that trade a clear benefit for richer, consented information: points balances, tier status, referrals made, and redemption history. Loyalty programs are powerful precisely because the value exchange is explicit, so customers willingly share more.
- Where it lives: loyalty platform, rewards system, referral tool
- Good for: high-intent segments, advocacy, repeat-purchase incentives
- Watch out for: a loyalty database that never merges back into the main customer record
9. Offline and in-person data
Offline and in-person data covers interactions that happen away from your website: in-store purchases, point-of-sale activity, event attendance, phone orders, and appointments. For any business with a physical footprint, bringing this back into the data foundation closes one of the largest blind spots there is.
- Where it lives: point of sale, booking system, event platform, paper and phone records
- Good for: a complete view across online and offline, local relevance
- Watch out for: offline activity that never gets tied to the same person's online profile
10. Contextual and device data
Contextual and device data is the situational layer: approximate location, device type, browser, and the channel a person is interacting through, all collected with consent. Used carefully it sharpens timing and relevance. Used carelessly it feels invasive, so it is the type to handle with the most restraint.
- Where it lives: first-party analytics, server logs, app SDKs
- Good for: timing, channel and device optimization, fraud signals
- Watch out for: collecting more than you can justify; apply data minimization here first
How the types map to outcomes
The types are not equally useful for every job. Knowing which type drives which outcome is what turns a data inventory into a plan:
- Acquisition and intent: behavioral and contextual data show who is leaning in right now
- Personalization: declared plus behavioral data tailors the experience without guessing
- Retention and churn: transactional, support, and account data flag risk early
- Lifetime value and expansion: transactional, loyalty, and account data find your best customers
- Compliance and trust: identity and consent records keep everything else usable
First-party vs the alternatives
All ten types above are first-party because you collected them directly. That is what separates them from second-party data, which is another organization's first-party data shared through a partnership, and from third-party data, which is aggregated and sold by companies with no direct relationship to the person. First-party types are the most accurate, the most compliant, and the most durable, which is why they are worth organizing properly.
The mistake: collecting in silos
Each type is useful alone, but the value compounds when they are unified to one customer record. A purchase plus a support ticket plus a stated preference tells a story that none of them tells by itself.
The most common failure is not under-collection. It is fragmentation. The email tool holds engagement data, the ecommerce platform holds transactions, the help desk holds support history, and the product database holds account data, but no system holds the whole person. As a result the same customer looks like four different people, and every team makes decisions on a quarter of the picture.
Connect them, do not just collect them
Unifying these scattered types into a single profile is the job of identity resolution, and it is what turns raw data into something you can actually act on. Once the types are connected, the next steps are to store them in a foundation you own and to activate them in the channels where they create value.
If you are not sure which types you already hold and which are trapped in silos, that is exactly what a Readiness Review maps out: where your first-party data lives today, what is connected, and what it would take to unify it.
Frequently asked questions
- What are the main types of first-party data?
- The ten common types are identity, behavioral, transactional, declared (zero-party), engagement, support and service, account and profile, loyalty and rewards, offline and in-person, and contextual and device data. Most businesses already generate all of these, often without realizing how much they collect.
- What is the most valuable type of first-party data?
- Transactional data is the highest-signal type you own because it reflects what people actually paid for, not just what they browsed. That said, the real value comes from connecting types together: a purchase plus a support ticket plus a stated preference tells a story none tells alone.
- Is an email list considered first-party data?
- Yes. An email list is identity data, one of the ten types of first-party data. But it is only a small slice of what you hold. Behavioral, transactional, support, and declared data all add context that an email address alone cannot provide.
- Where does first-party data live?
- It is usually scattered across the tools that collected it: identity in the CRM, behavior in analytics, transactions in the ecommerce or billing system, support history in the help desk, and declared data in survey or preference tools. The work is connecting these silos to one customer record through identity resolution.
Know what first-party data you already own
Get a free First-Party Data Readiness Review to map where your data lives today, or score yourself in minutes with the readiness checklist.