Leave a comment

Detailed guide to use Google Tag Manager (GTM) to personalise your website. 100% free! :)

Learn how to use GTM to:

  • Create Split Groups & track them in Google Analytics to measure A/B tests
  • Step-by-step guide how to use Intent-based Personalisation on any Website to change Headlines & Copy, Images etc. (e.g. using visitor landing pages, UTM codes)
  • Advanced Sumo.Me tricks to target based on dataLayer variables and more! All non-coder friendly & w/ tag manager containers :)

Please post comments here or on the blog for feedback, questions or how you've used it!

  • GG

    Gregory Gerla

    9 months ago #

    This is a great introduction to rules-based personalization. Makes total sense to have 'when-x-do-y'.

    I think the problem we are running into as an industry is that it get's too complicated at scale. You have to have someone actively updating the tags (front end dev), hypothesizing new ideas (marketing), analyzing results (data science), which is all costly. At a big B2B site, there are tons of moving pieces, and I don't know how this scales to the enterprise. Of course, this is growth hackers, so this might be perfectly suitable on a simple site.

    Additionally, there is a problem I see as it relates to the distribution of traffic to value.

    This post makes sense when there is an equal distribution of value across your audience, EG: B2C.

    However, what happens if only 5% of your traffic makes up 95% of your revenue? You might over-emphasize certain tests thinking they are converting better. The truth might be the opposite, where that 5% is part of a losing test. The solution then is to consider down-funnel goals, such as revenue, not just conversions. Alternatively, if you knew more about your traffic, you should weight the test accordingly.

    Something to consider when deciding if spending the time on this is worth the effort.

    • AG

      Alem Galvez Alcantar

      9 months ago #

      I couldn't agree more. Well said.

    • JR

      Johannes Radig

      8 months ago #

      Hi Gregory - you're absolutely right. Using GTM for this purpose is probably not a solution forever; but it's a good way to build an MVP or 'proof of concept' and then iterate from there. After all, this is crucial for any 'growth' experiment - and it helps to justify the budget for some of the powerful tools out there.

      Thanks for your comment!

      Best
      Johannes

    • JR

      Johannes Radig

      8 months ago #

      To add to this bit:

      "However, what happens if only 5% of your traffic makes up 95% of your revenue? You might over-emphasize certain tests thinking they are converting better. The truth might be the opposite, where that 5% is part of a losing test. The solution then is to consider down-funnel goals, such as revenue, not just conversions. Alternatively, if you knew more about your traffic, you should weight the test accordingly."

      Agreed - but there are things we can do to cater for this 5% a lot better then with for example a generic homepage. Evaluating test results should never always be done by looking at overall conversion rates. But Google Analytics allows analysing CR% for more specific segments / dimensions, and different goals. Totally crucial to consider this. Thanks again, shame I only saw your comments now :)

Join over 70,000 growth pros from companies like Uber, Pinterest & Twitter

Get Weekly Top Posts
High five! You’re in.
SHARE
31
31