Google Publisher Tag (GPT) is an ad tag library that allows publishers to define inventory, initiate and bundle ad requests, and render matching demand. GPT takes key details from you (such as ad unit code, ad size, and key-values), builds the request, and displays the ad on web pages.
With GPT, you can define shared settings, such as targeting at a page level, instead of defining the same targeting for each individual ad on the page. You can also control how and when ad requests are made. For example, you can use single-request architecture to call all ads once in the header of your page.
The JavaScript files which implement GPT are served by Google. You should not serve these libraries from your own server or load them from an unofficial source.
Get started with the developer documentation
How GPT works
Google Publisher Tag is used to define available ad slots in your network. Placing GPT on a page creates a communication path between the ad server and a user’s browser. When a page containing GPT is requested, the following sequence of events occurs:
-
A request is made from the user's browser to the ad server for
gpt.js, the tag code built in JavaScript. -
The JavaScript builds and sends one or more requests to the ad server for ads tagged on the page.
-
The ad server matches the ad units and any key-values contained within the request.
-
The ad server runs an auction to select the best matching creative for the ad unit.
-
The ad server returns the matching ad to the GPT JavaScript.
-
The GPT JavaScript renders the ad on the page.
Benefits of GPT
There are multiple benefits to GPT:
- Multi-level inventory hierarchy: Google Publisher Tag allows you to use more granular levels of inventory in the Ad Manager front-end. With multiple levels of hierarchy, you can create much more specific targeting based on your site content.
- Minimizes page load time impact: An asynchronous JavaScript fetch means your webpage renders without having to wait for the JavaScript to be returned from the Ad Manager servers. Ads load into iframes as soon as the creatives are returned from the server.
- Google Publisher Console: The tag comes with a built-in debugging and support tool called the Google Publisher Console, available on all pages containing the Google Publisher Tag.
- Single request architecture (SRA): Instead of sending individual ad requests to Ad Manager servers, the tag is able to send one request notifying the server of all ad units on the page. This is the recommended request mode, as it allows advanced roadblocking and improves page load time.
- SafeFrame capability: SafeFrame technology allows transparent interactions between page content and ads, while providing publisher control and protecting you from potentially untrustworthy creatives and auto-redirects.
- Responsive design: You can use GPT to build ads that fit the browser your visitors use to view your website. This means that your ads look good on any device.
- Page-level ad selection: GPT and the ad server work together to keep track of what ads are on the page, so you can define roadblocks, takeovers, and competitive exclusions.
- Support reactive ads: You can use GPT to build and display web interstitial, anchor, and rewarded ads formats.
- Manage Consent Framework integration for compliance: GPT can communicate with IAB CMPs to handle and respect users' consent regarding storage usage, and personalization. GPT also provides APIs for publishers to configure privacy settings and limited ads.
GPT and secure pages (https://)
GPT works automatically with secure webpages whose URLs begin with https://. There's no need to modify the tag in any way for serving on a secure page.
HTTPS stands for HyperText Transfer Protocol Secure. It's a communications protocol built on top of the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols. These protocols encrypt certain elements of the communication between the server and the client to prevent tampering and eavesdropping by third parties.
