How to implement Google Analytics Anyway?

Most of the companies I worked with are Tech StartUps where Analytics impementation will be last thing on the priority list for Tech Team.

I don't blame them, as a tech startup they are suppose to make drastic changes to the whole site with the blink of an eye. On the other hand, there are corporate firms or rather large startups who has developers specifically assigned for analytics implementation.

There are three ways by which one can implement Google Analytics based on the availability of tech resource. Will discuss the three ways in detail in subsequent posts in this series.For now, let's look at the pros and cons of all three, and decide which one is best suited for your firm.

1. T-Team

You are one of the luckiest Web Analyst in the World.  

There is a team of developers dedicated to Analytic Implementation.

With the help of this team, every tag and trigger of Google Analytics is hard coded on your site.

And since the condition associated is actually a part of your site code, the chances of data spill is close to ZERO. This is actually the ideal type of analytics implementation in terms of minimum data loss during collection. 

But this comes with a cost, every time you need to create or change any tag or trigger, your team has to make changes in the website code and you have to wait for next website release cycle which depending on companies varies from days to months.

Pros - Minimum Data Loss 

Accurate and to the condition triggers

Cons -  High IT dependency. Every small change might require a web release

2. Lone Ranger:

When there is no way, Google provides you one, Google Tag Manager (GTM) you beauty!

In GTM you can write your own triggers and tags based on the front end elements.

But you must have some basic knowledge of the front end divs, easily acquired with an hour of discussion with your front end developer. This includes you to understand how to get values of required elements from the document objects of your site. If you additionally have some basic JavaScript knowledge then you are in for a treat, as GTM gives you power to write your own Custom HTML which you can use to create JavaScript code snippets and use them in your current site, without any web release.

But since these tags and triggers' definition is based on front end elements, you need to keep track of any front end changes on the website and update your tags and triggers accordingly.

Pros - No Dependency on tech team or website release cycle.

Cons - Have to keep track of all front end changes.

3. Middle Ground:

This case is the combination of first two cases, with some dependency on tech team but dedicated tech developer is not required.

Here we will be using GTM and DataLayer. A DataLayer is a JavaScript array in which data required by GTM, is stored. As an analyst GTM ownership will be yours and that of DataLayer will be of developer's. Tags and triggers will be created on the basis of this DataLayer. I generally ask my developer to send all the relevant information to DataLayer, which currently may not be of use but may be useful in future to create tags and triggers, this eliminates my dependency on tech. And since DataLayer is getting generated in the code itself so no need to worry about any front end changes.

Pros - No dependency on tech for small changes

More data accuracy than 2nd Case

Cons- large Changes may need tech team's intervention

Conclusion : 

Your Google Analytics implementation will fill in one of the above category, as I mentioned it depends on the Business needs which path you want to choose. Analytics Implementation is not a one time task, it is an ongoing process, keeping this in mind  I suggest discuss it with all stakeholders, including Tech and then decide which direction to take.