Data broker glossary
A data broker glossary of the terms you meet when companies buy, sell and license data, each explained with a worked example.
Updated
All glossary
| Page | What it covers | Updated |
|---|---|---|
| What is DaaS (data as a service)? | DaaS, short for data as a service, means a provider keeps data current and customers reach it on demand instead of buying a static file. | 2026-11-07 |
| What is an AI dataset? | An AI dataset is the material a model learns from: what it contains, how it is organised, and what buyers check before licensing one. | 2026-11-07 |
| Data enrichment: what it means | Data enrichment adds context a dataset lacks, from outside sources or from analysis, so it answers more questions than it could alone. | 2026-11-07 |
Why a glossary for data deals
The trade in data borrows its vocabulary from marketing, law, finance and machine learning at once, and each field uses the same words a little differently. A marketer hears "first-party data" and thinks of cookies and consent banners. A lawyer hears it and thinks about who collected the information and under which privacy notice. An AI lab hears it and wonders whether the seller can prove where it came from. When a seller and a buyer use one word in two senses, the misunderstanding usually surfaces late, in diligence or in the licence, where it is expensive to fix.
Every entry here starts with a short definition you can quote, then shows the term at work in a realistic dataset deal and sets it beside the words it is most often confused with. Where a term has a legal meaning in a state law, the entry says so and points to the law page. Where an entry mentions a figure, it names the source; the rest is explanation, not data.
Every term, A to Z
| Term | Where it comes up |
|---|---|
| AI dataset | Matching: what AI buyers want |
| Alternative data | Matching: investment buyers |
| B2B data | Sourcing: company and contact data |
| Consumer data | Qualifying: personal data and the law |
| Data aggregator | Sourcing: who combines sources |
| Data append | Valuing: filling gaps in a file |
| Data as a product | Sourcing: how sellers organise data |
| Data as a service (DaaS) | Closing: delivery by subscription |
| Data broker | Every stage: the role itself |
| Data clean room | Closing: sharing without handing over |
| Data commercialization | Sourcing: bringing data to market |
| Data enrichment | Valuing: adding context to records |
| Data licensing | Closing: the legal instrument |
| Data marketplace | Matching: platforms that list data |
| Data monetization | Sourcing: why companies sell |
| Data ownership | Qualifying: who may license |
| Data product | Valuing: the packaged dataset |
| First-party data | Qualifying: where data came from |
| Information broker | Every stage: the older name for the role |
| Intent data | Matching: sales and marketing buyers |
| LLM training data | Matching: language model buyers |
| Second-party data | Sourcing: partner data |
| Third-party data | Qualifying: bought-in data |
| Virtual data room | Closing: diligence and samples |
| Zero-party data | Qualifying: data people chose to share |
Terms that travel in groups
- Zero-, first-, second- and third-party data describe one question from four angles: how far the data has travelled from the people it describes. Read them together.
- Monetization, commercialization and licensing are often swapped. Monetization is the goal, commercialization is the work of taking data to market, and licensing is the contract that lets someone else use it.
- Data product and data as a product sound alike but aren't. The first is a thing a buyer can license; the second is a way of running data inside a company.
- Enrichment and append both add to a file. Append fills named gaps by matching; enrichment is the wider job of adding context.
- Data broker, information broker and data aggregator overlap heavily. Each entry says where the lines usually fall, and the law pages show where a statute draws them.
How the entries are written
Examples are illustrations, written to show how a term behaves in a deal, not reports of real transactions. Definitions are our own wording. Legal points describe the law in general and say when to ask a lawyer. Each page shows the date it last changed and lists the sources behind any factual claim with the date they were checked.
Questions people ask
What is the difference between a data broker and an information broker?
In everyday use they overlap. Information broker is the older term, often for researchers who find information on request; data broker is the term state laws now use for businesses that sell personal data about people they don't deal with directly.
Which terms matter most to someone new to brokering datasets?
Start with data broker, data licensing, data ownership and first-party data. Together they cover what you do, the contract you use, who may sign it and where the data came from.
Are these definitions legal definitions?
No. They explain how the words are used in practice. Where a state law gives a term a specific meaning, such as data broker, the entry links to the law page, and the statute's wording is what counts.
Related
Become a data broker
The DataBroker course and tools: find companies with data to sell, price it, license it and close the deal.