The ad tech glossary

309 terms, each answered in one sentence, then explained properly with an analogy and a worked example.

All Basics Programmatic Invalid traffic & fraud Measurement Data & identity Privacy & regulation CTV & video Mobile & app Retail & commerce Formats & creative Business & finance Search & social Supply chain Audio & DOOH

A

Audience segment: An audience segment is a group of people or devices defined by shared characteristics, such as demographics, interests, purchase behaviour or intent, used to target or measure advertising.

B

Behavioral targeting: Behavioral targeting shows ads to people based on their past online behaviour, such as sites visited, searches, purchases or app use, collected over time and often across different services.

C

CDP (customer data platform): A CDP (customer data platform) is software that collects a company's first-party customer data from many sources, unifies it into individual profiles and makes those profiles available to marketing tools.
Contextual targeting: Contextual targeting places ads based on the content being viewed, such as the topic, keywords or sentiment of a page, video or app, rather than on data about the individual viewer.
Cookie syncing: Cookie syncing is how two ad tech companies match their separate cookie IDs for the same browser, so they recognise the same user when exchanging bid requests or data.
Cookieless advertising: Cookieless advertising means targeting and measuring ads without third-party cookies, using approaches like first-party data, contextual targeting, alternative IDs, clean rooms and modelled measurement.

D

Data broker: A data broker is a company that collects personal information about consumers from many sources, often without a direct relationship with them, and sells or licenses it to others.
Data clean room: A data clean room is a secure environment where companies combine and analyse their customer data together, getting aggregated insights or audiences without exposing each other's raw personal data.
Deterministic matching: Deterministic matching links identifiers to the same person using definite, shared information, such as the same login email or phone number, rather than statistical inference.
DMP (data management platform): A DMP (data management platform) is software that collects and organises mostly anonymous, cookie-based audience data from first and third parties into segments for targeting programmatic ads.

E

European Unified ID (EUID): European Unified ID (EUID) is the European version of Unified ID 2.0, an email-based, hashed and encrypted advertising identifier designed to work within GDPR's consent requirements.

F

First-party data: First-party data is information a company collects directly from its own customers or audience, such as purchases, sign-ups, app use and site behaviour, through its own channels.

G

GAID (Android advertising ID): GAID, the Android advertising ID, is Google's user-resettable identifier provided by Google Play services that Android apps can use for advertising and analytics.
Geotargeting: Geotargeting is showing ads to people based on their geographic location, from country and city down to postcode or a small radius around a store, using IP addresses, GPS or declared location.

H

Hashed email: A hashed email is an email address transformed by a one-way cryptographic function, usually SHA-256, into a fixed string used to match users across systems without sharing the plain address.

I

ID bridging: ID bridging is attaching an identifier observed in one context, such as a hashed email from a logged-in session, to impressions in other browsers or devices where it was not directly observed.
Identity graph: An identity graph is a database that links the many identifiers belonging to the same person or household, such as emails, device IDs, cookies and IP addresses, into one profile.
IDFA: IDFA (Identifier for Advertisers) is Apple's resettable advertising identifier for iPhones and iPads, which apps can access only if the user grants permission through App Tracking Transparency.
IP address (in advertising): In advertising, an IP address is the network address a device uses online, used for rough geolocation, household matching, frequency capping and fraud detection, and treated as personal data in many laws.

L

Lookalike audience: A lookalike audience is a group of new people who statistically resemble an advertiser's existing customers or best users, found by a model to extend reach beyond known contacts.

M

Mobile advertising ID (MAID): A mobile advertising ID (MAID) is a resettable identifier that a phone's operating system provides for advertising, such as Apple's IDFA or Google's GAID, letting apps recognise the device for ads.

P

Privacy Sandbox: Privacy Sandbox was Google's initiative, launched in 2019, to build browser and Android APIs to replace cross-site tracking; in October 2025 Google announced it would retire most of those APIs.
Probabilistic matching: Probabilistic matching links devices or identifiers to the same person or household using statistical inference from signals like shared IP address, location, timing and device traits, with some uncertainty.
Protected Audience API: The Protected Audience API (formerly FLEDGE) was a Privacy Sandbox proposal to run remarketing ad auctions inside the browser without cross-site tracking; Google announced its retirement in October 2025.

R

RampID: RampID is LiveRamp's pseudonymous, people-based identifier, which links a person's or household's online and offline identifiers so advertisers and publishers can match audiences without sharing raw personal data.

S

Second-party data: Second-party data is another company's first-party data shared directly with you through a partnership, such as an airline sharing audience insights with a hotel chain.
Seller-defined audiences: Seller-defined audiences (SDA) is an IAB Tech Lab specification that lets publishers label ad requests with standardised audience segment IDs from their own data, without passing personal identifiers to buyers.
Signal loss: Signal loss is the reduction in data available to target, measure and optimise advertising, caused by browser cookie blocking, mobile ID restrictions, privacy laws, consent choices and ad blocking.

T

Third-party cookie: A third-party cookie is a small file set in your browser by a domain other than the site you are visiting, typically ad tech, letting it recognise your browser across sites.
Third-party data: Third-party data is information about people collected by companies with no direct relationship to them, then aggregated and sold or licensed to advertisers, often as ready-made audience segments.
Topics API: The Topics API was a Google Privacy Sandbox feature that shared a few broad interests, inferred from recent browsing, with sites for ad targeting; Google announced its retirement in October 2025.

U

Unified ID 2.0 (UID2): Unified ID 2.0 (UID2) is an open-source advertising identifier, originally developed by The Trade Desk, generated from a user's hashed and salted email address or phone number with their consent.
User agent: A user agent is the text string a browser or app sends with each request describing its browser, version, operating system and device, used for compatibility, analytics and fraud detection.
User-Agent Client Hints: User-Agent Client Hints are a web standard, championed by Chrome, where browsers share only basic details by default and send detailed device and browser information only when a site asks.

Z

Zero-party data: Zero-party data is information a customer intentionally and proactively gives a brand, such as preferences, intentions or quiz answers, rather than data inferred from their behaviour.