The Empathic Analyst

When most people open Google Analytics or any analytics system for that matter, the first thing they see is… 

Numbers and graphs. 

Row upon row, column after column, line after line. 

And when most people see numbers and graphs, they think math and science. 

when most people see numbers and graphs, they think math and science

Do we agree so far?

Because if we agree on this we have potentially identified one of the most critical issues with the perception of digital analytics.

What makes our digital analytics data unique is that it is not scientific and conclusive. For years we have been talking about how wrong the data is and how many issues there are with the quality. 

Still we have been trained that what we are doing is serious statistics and hard science. 

But the problem is that what we are doing is not always what we think we are doing and it is often less difficult than we make it. 

Because honestly, it is not rocket science. 

it is not rocket science

It’s is actually not even science… Well, not like we like to think it is. 

Because we work with numbers there has been a massive tendency in trying to put what we do in the same category as math, statistics and the natural sciences. Areas with universal fundamental laws, which would stand the rigor of angry scientists using the full range of rules and tools to qualify something as new universal knowledge. 

And guess what. 

analytics does not qualify as science

96.23 % of the activities done with and in digital analytics does not qualify as science or even facts for that matter. 

Why? 

Simple really.

In most cases we are not working with the fixed forces of the universe, we are working with people. Behind all the code, tracking, numbers, graphs and metrics it’s basically just people in the real world. 

That means our results are impacted by the mood of the audience we serve it to and since they again are impacted by time, content and context our results very often will only have value on a short to mid-term term basis.  

our results very often will only have value on a short to mid-term term basis

And yet we have been spending time thinking we have been searching for some universal greater truths.

That the green button always will be better than the red, that the best time to post on Linkedin is on Tuesdays at 11.00. 

But people change their minds, fashion changes and technology moves.  

So the answer to the data in analytics might not be science, it’s certainly both math and statistics… and most likely also a lot empathy. 

The data represents the behaviour and patterns of people and how they act, interact, react or simply dont around our digital activities. 

So we need to recalibrate our numbers to people.  

As one of the signs of a great analyst Avinash Kaushik points out that you should visit the page that you are analyzing, to see the numbers through the eyes of the users. 

And it makes sense. 

our job is to secure the return on marketing

If our job is to secure the return on marketing, web and digital then we need to focus on what actually generates that return and for digital analytics that is the end-users. 

So once the numbers have been added the conclusion is not that the trend is an increase in the conversion rate. The trend is that more people conclude their transaction on your website.

And the numbers will not tell us why, but the site and communication should. 

By distancing ourselves from the users we miss the biggest value which can be added to the analysis, the considerations around the customers.

Data driven is not customer centric

Data driven is not customer centric until we fully accept and absorb that the digital analytics data represents the people.  

Because at the end of the day we are not changing numbers, we are moving people and people respond to communication. 

So maybe instead of only giving the numbers to a mathematician or a statistician we should serve them up for a psychologist?

we need to try to understand them through empathy and not math alone

The moment we accept the numbers reflect people then we need to try to understand them through empathy and not math alone. Why people act is often more an emotional question than a logical one and any attempt to explain behavior only with logic will rarely uncover the full story. 

You can rarely read how confused, angry, delighted or annoyed I am by looking at my aggregated data.  

In that sense data is not one thing and all data is absolutely not created equal. 

all data is absolutely not created equal

Because the double digit precision we serve up our metrics and measurements with is an illusion. For a thing to qualify as math or science it needs to be repeatable and systematic over time. 

Few of our measurements are scientific facts. No matter what wizard sets up your tracking setup the majority of the number will be a reflection, not of facts, but of human behavior. 

So the most important part of the analysis is to put away your calculator for a minute and take a look with your heart. 

put away your calculator for a minute and take a look with your heart.