A tag manager is a tool that lets a website owner add, change and remove scripts, called tags, such as analytics and advertising pixels, from one interface. The page includes a short snippet that loads a container, and the tags inside the container are changed without editing the site's code.
Parts of a tag manager
- Container: the bundle of tags, triggers and variables that a page loads.
- Tag: a piece of code, such as an analytics or advertising tag, that runs on the page.
- Trigger: the rule that decides when a tag fires, such as on page load or on a click.
- Variable: a value that a tag or trigger uses, such as the page address.
Google Tag Manager is one example, with containers identified by an ID that begins with GTM-.
Why it matters for technology detection
A tool loaded through a container does not appear in the page's HTML, so a detector that reads only the page understates a company's technographic data. A complete detector reads the container as well: every visitor's browser downloads it, so the tags inside it can be read from it, alongside the scripts, network requests, response headers and DNS records of the site. Three behaviours of tag managers matter:
- A tag can be set to fire only after consent, only on certain pages or only under set conditions, so a full picture combines several views of the site.
- A container can add or remove several tools at once with no change to the site's own code.
- Tags come and go with consent banners and tests, so a removal is established over more than one observation, not a single one.
In Fokals data
Technology Stack and Technology Changes include the technologies deployed through tag managers, each with its first-seen and last-seen dates, and a dated event for every adoption, removal and new account ID under a tool already seen. The methodology and the marketing stack dataset give the detail, and the article on how Fokals detects technologies walks through it.
Related terms
- Installed base: the set of customers running a vendor's technology, which tag detection helps to count.
- Technology churn: the rate at which companies drop tools, which depends on reliable removal rules.
- Data coverage: which companies and technologies a dataset includes.
Frequently asked questions
What does a tag manager do?
It lets a site owner deploy and change third-party scripts, such as analytics and advertising tags, without a developer editing and releasing the site's code each time. The site includes one container. Marketers set up tags, triggers and variables in the manager's interface and publish a new version of the container.
Can you see which tags a site loads through its tag manager?
Yes. The container is a script that every visitor's browser downloads, so its tags can be read from it, as can the network requests a page makes. Tags set to fire only after consent, on other pages or under set conditions depend on those conditions, so a complete picture combines page content, scripts, network requests, response headers and DNS records.
Why does a technology appear and disappear on a website?
Tags come and go with consent banners, tests and container changes, so a tool can be absent from one reading and present in the next. To avoid false removals, treat a tool as removed only when it stays absent on a later reading some hours on, and check that the view it was seen in, such as its container, could be read.