The global site tag (gtag.js) is now the Google tag. Learn more

Upgrade guide

TODO: This will be the upgrade guide based on this sheet.

To get the latest functionality from Google measurement products as they become available, Google recommends that you deploy measurement tags via gtag.js or Google Tag Manager.

Legacy tag platforms such as conversions.js and analytics.js can upgrade to gtag.js or Google Tag Manager. To make sure your upgrade process goes as smoothly as possible, consult guidance for each specific product.

Not sure what tag platform you currently use? Read our guide on how to analyze existing tags!

If you use gtag.js

In most cases you are using the latest tag library. Good job! In some cases you may need to update your instrumentation to use common event definitions that work across your products and interoperate with the latest reports.

If you use Google Tag Manager

Check to see which versions of tags you have deployed with Google Tag Manager. If you are using an older tag platform (e.g. Universal Analytics) you can upgradce to a more advanced version with a few simple clicks. Simply add the updated version of your tag, apply appropriate trigger conditions for your tag, configure instrumentation (such as ecommerce), and publish your container for the changes to take effect.

If you use legacy tag types

If you have legacy tag types installed, such as ga.js, analytics.js, or conversion.js, you should consider an upgrade to gtag.js or Google Tag Manager.

If you use a third-party Tag Management System

If you use a third-party tag management system (TMS), check to see if the TMS uses the latest version of Google's tag platform. If not, you may need to update your tag template or custom code to get the latest functionality.

If you use a mix of tag platforms

We know how it is. In the real world, as sites and apps grow and change management, tag technologies have a way of evolving.

Optimally, your site will use one single tag platform. Consider consolidating your tags to the single platform if possible. This will help reduce errors in data collection, improve performance, minimize the chance of bugs appearing, and simplify ongoing management.