I’m a marketing professional who has been doing basic tagging and analytics for clients. I’ve encountered an account set up by someone with extensive development/analytics knowledge, and I’m struggling to understand some of the custom tags and variables. Specifically:
- What is a ‘GTAG Get API – Client ID’ tag? How does it differ from a standard Client ID tag?
- There’s a tag called ‘GTAG Client ID’ that seems to use a custom template. What might this be doing?
- For conversion tracking, there’s a tag named ‘onboard step tracking_2024’ with a variable containing JavaScript. What could be the purpose of this more advanced setup?
I may need to replicate these tags for other clients, but I’m unfamiliar with these more complex configurations. Can anyone with experience in advanced GTM setups explain what these tags might be doing and why they might be used instead of standard implementations?
Any insights or explanations would be greatly appreciated. Thanks!
I have played around in GTM and preview mode, as well as inspecting the page. That is how I have found the detail I have. This page also experiences a below normal bounce rate at 3% on average, so I know something else might be off. It is hard for me to decipher with all of this javascript involved.
I also found that A GTAG Get API – Client ID tag in Google Tag Manager (GTM) is a custom tag template that allows you to retrieve the Google Analytics client ID and other fields using the gtag.js API. It’s a custom tag template created by Simo Ahava that can be added to your GTM workspace from the template gallery. HOWEVER, this doesn’t help me much, just gives origination context.
See my preview/debug link here: https://tagassistant.google.com/#/?id=GTM-NKPL4LG&url=https%3A%2F%2Fapply.zynlobank.com%2F%3F_ga%3D2.228588379.717648131.1726682440-1476769114.1724684401%23%2Fintake%2Fobject%2Frao%3Ffield%255Bstep%255D%3D2%26field%255Breferral%255D%3DTWO&source=TAG_MANAGER>m_auth=KfdaAJLWOmUYMVQgwn07Tw>m_preview=env-3
Sophie is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
-
Judging from the name, that’s not a tag but a variable. And from the name of it, it seems like it depends on Simo’s template that uses gtag API to pull its current state variables and dump them into the DL. The variable then picks the client id from there. When you say the standard client id, I don’t know what you mean. I just checked my GTM and there’s no such variable. Or tag.
-
Open the template, see what it does. It’s all open source.
-
More advanced? Using JS is pretty normal in a TMS. A martech developer has to be comfy with JS. At least the basic things. Again, hard to tell without seeing the code, but there are many valid reasons to have JS in a TMS, especially when there is not enough front-end resources to own all the data molding logic on their side.
To debug anomalous bounce rate, you don’t need go to JS. You just need to know how bounce rate works and how to look at the GA events firing. You install an extension or go to network and look at your ‘collect’ calls. In them, you need to look at the interactional calls. The property in the body of the request for them would be ni. If there are interactional hits after the initial pageview, the visit won’t bounce.
In conclusion: you need to know JS to do more advanced TMS setups. It’s very rare that a TMS setup won’t require JS knowledge. And that would normally be a very unreliable or simplistic setup.
It’s possible to create reliable and complex TMS setups with no code, but it’s rare. Requires quite an effort from everyone involved: PMs, scrum teams, analytics engineers. That’s usually not the case even though many think that’s how it should be done.