Google Tag Manager
- area
- Measurement
Google Tag Manager is the mail room of your website. Your page drops a note in the tray every time someone views a page, adds to a cart or fills in a form, and Tag Manager reads the rules you wrote and decides which vendors get a copy. Technically, it is one JavaScript file, gtm.js, that your site downloads on every visit, carrying every tag, trigger and variable you have ever published. Change a rule in the interface and you have changed your website, without touching its code.
That is why I like it, and why it worries me. A container that has been around for a few years tends to collect things. Three consent tags, and nobody is sure which one is live. A Custom HTML tag called "test", which has quietly fired on every page since before anyone can remember. None of this is anybody's fault. It is what happens when a tool makes adding things easy and removing things frightening.
What I do with it:
- Audit and clean up. I read the whole container, work out what each tag actually sends and to whom, and remove what nobody can explain.
- Consent that works. Consent mode, set up so that tags wait for a visitor's choice and respect it. A German court ruled in 2025 that Tag Manager itself can need consent before it loads, so this is a legal question as much as a tidy one.
- Server-side tagging. A second container on a server you control, so your site sends one request and you decide what each vendor receives. Fewer scripts in the browser, and far fewer surprises.
- Rebuilds. For a container past saving, a clean one with a naming scheme, documented rules and a test plan, so the next person does not inherit a mystery.
The newest thing I have written about it is Tag, You're It: Google Tag and GTM Become One Container, about Google's August 2026 change and whether to click the blue button. It is long. Make a pot, not a cup.
Alongside it
The platforms it usually turns up with. They rarely travel alone.
