Anyone self-hosted the js tracker just because the Tag Manager method is long and has limitations

I don’t know why it has to be that hard, the setup of the dataLayer, and the limitations, why would anyone use the Tag Manager?

Our recommended method is to self host and to implement it directly into your application, this offers the greatest amount of control in how you track your events.

However, I assume you’re using Google Tag Manager - are you using the GTM Snowplow Template? If so, what are the issues you are facing? We’re always open to feedback.

Yes, we are selfhosting js tracker. The main reason is:

  • GTM is blocked by many adblockers
  • You need one more request so it’s a bit slower

The only situation we use GTM is the when we want to duplicate Google analytics events to Snowplow.

Hi Paul,

So my experience was:
I was doing the installation on GCP, and it was not easy. Then I came to this page, which says:

We strongly advise new Snowplow users who are not using a Tag Management solution to implement one before implementing Snowplow, and integrate Snowplow using it.

So I decided to use GTM, then I came here and found out that there are some caveats and the installation is long, which unusual for a script you just want to put in the page.

Actually, I am thinking of doing everything from the server to avoid adblockers.

Thats the best option for sure, rename sp.js too.

You’ve stumbled across an old docs page there. We’re migrating to https://docs.snowplowanalytics.com now but still have a few remaining bits and bobs we need to tidy up on GitHub. I’ll sort that page in particular out now, the rest of the JavaScript docs will make the final move when we release v3 of the tracker.

As for GTM, you can do it either way. You can use the Custom Template or you can add a Javascript snippet using GTM.

Thanks Paul