Tuesday 29 October 2013

Google Analytics, Marketing Attribution and the True Football Analyst:

Now, it would probably amaze some of you that I want to relate web analytics with football analysis! Well, I am doing so because I live in Nigeria in Africa and there isn't that much "analyzing of websites" going on in this part of the world, I can tell you that much! And this is mainly because most business owners haven't yet come to the realization that they need to analyze their effectiveness. 

So what I am going to do with this article, is try to relate this concept of Analytics to another one that practically all African people (Nigerians especially) know so well....


Now, I would first need to quickly point out that I will be discussing soccer, and not the NFL. In short, European football and not American Football. Okay that wasn’t shorter, but I'm sure you get the point.
I must also mention that, unlike almost every soccer fan I’ve met, I am the very rare type that actually hates to “analyze” why a team won or lost a football match after the 90 minutes have been completed. So often times I say things that shock practically everyone that call themselves football fans.

For example, I still insist that Diego Maradonna’s Argentina squad in 2010 was tactically one of the best since 1994. In fact I consider Jose peckermann's 2006 squad the only one better in that time frame. A claim that not one human being I know agrees with.
But I don’t care about that; it’s not my problem….they are the ones who need to open their eyes. Hahaha!

Anyway, seriously, you should also know that I never judge the performance of a soccer team according to the number of goals that were counted for or against them at the end of one single game. So my take is that unpredictable things happen during the course of 90 minutes that sometimes make the beautiful game of football seem to resemble an adulterous whore. Do pardon the illustration.

But now let me quickly get back to this discussion and define Marketing Attribution and Google Analytics so that the football fans reading this article can catch up with what I am about to introduce to you geeks out there:


Google Analytics is a powerful instrument that is used for measuring a website’s activities, like visits, clicks, and the defined goals that may have been setup by the website owner.
These activities need to be measured when you decide to start using your website as a marketing, branding or promotional instrument for your business or whatever cause the website was built for.

So burn this in: A website shouldn't be just a dummy in cyberspace that most companies around you have one of and so you feel you must therefore have one yourself.
You’ve completely missed the point, if you go this way.

Visits and Clicks to your website can come from any part of Africa or the world at large. They can be attracted by your brand name, or they can be attracted by your website’s marketing strategies and tactics. Either way, once they’ve come to your website, chances are, you can get them to buy your products, visit your shop or direct their colleagues to it, subscribe to your articles, and even promote your brand for you. It all depends on how you much you are willing to get out of digital marketing.

Conversions are the completion of goals that are specified by the website’s owner as important things he would like visitors to do while on his site. These goals will depend completely on whatever is the nature of your business, and they will be the major things that you would be measuring with Google Analytics. Additionally, if you have an e-commerce website, these goals would primarily be the actual purchases of items made by online customers on your site.

So Google Analytics is like an instrument of forensic examination that helps you scrutinize your website’s performance in terms of how well it is achieving its GOALS (just like in soccer you know?). It is the only way to know for sure, what and where you need to make improvements so that your website can perk-up on its performance and begin to achieve those goals, and reach for greater goals.

But the aspect of Google Analytics that I really wish to discuss today is Marketing Attribution. 

Along with Advanced Event Tracking, Attribution is one of the least mentioned topics in GA by the so-called digital marketing "experts" in Nigeria. I have also searched online for articles on this topic by other digital marketing professionals from the African continent; sadly, I've found none.

Nevertheless, the impact of Attribution (when it is well done) on your website and your entire business could be quite useful in creating quite a mismatch between you and your competition on the digital frontier.

So without further ado, Attribution is about giving value to different channels within your overall marketing campaign so that you can get a clearer understanding of how effective each of the various channels or “players” in your marketing efforts are performing.

We use Attribution Modeling because often times digital marketers think that one aspect of their marketing strategy is working better than other channels, and as a result they tend to focus so much effort on that particular medium at the detriment of other channels. The potential problem here is that if you are mistaken with this assumption, then your marketing campaigns may never reach their full potential.


So let's go back to football for a while:

Now, if you were to take away both Mezut Ozil and Angel DiMaria from Real Madrid FC, you could be surprised that Cristiano Ronaldo would score fewer goals than you know him to score. 
I could have said the same about Iniesta and Xavi on Leonel Messi, but I have seen otherwise.

Anyway I don’t mean to start an argument here, I am only pointing out that it is important for you to learn how credit should be shared in your marketing campaigns. And thankfully, your marketing campaigns don’t have huge egos. Hahaha!

And for those who wish to argue football, you can join my football community on Google+. But you should be warned that there are many who consider me a real rabble-rouser, who became an obsessive football watcher when I was 16 yrs old.

I had started disagreeing vehemently with the usual British commentators and studio pundits from when I was about 20, and since then I have very seldom ever taken their kind of football analyses seriously.

By the time I was 22, I was watching some matches 3 to 5 times over, just so that I could discover patterns in certain formations. It even got to a point where I preferred watching a recorded game to a live one; since in a recorded match, you may easily press review to understand why a player made a certain move (whether it was a fluke or a calculated knack of genius).

In any event, the relevance of all I am saying is that (and this is just my opinion) Attribution in Google Analytics also requires a similar kind of obsessive behavior in paying attention to analytics data. When I first started using GA, I thought it was extremely complex and boring. But the trick was to tarry and then learn first to pay attention to the simpler details before graduating to the more advanced metrics. 

If you did that, and also opted to follow the most knowledgeable experts and the insights on Attribution that they very generously share on their blogs, You would suddenly realize that this could all be so much FUN………Or maybe not.


Anyway that depends on the kind of person you are, I guess.
For me the connection became obvious when I first noticed that the Last Click assumption that almost every digital marketer was pre-occupied with, could easily be likened to the Leading Goal-scorer in a football team, and how most people assume that he scores those goals by conjuring them all out of nothing. So just like in football, most people come to digital marketing with the Last Interaction model of Attribution already ingrained in their minds.
In case you are a bit confused with what I am talking about, I will give you a quick background of how it works.

Marketing Attribution in Google Analytics, comes with a few models that are given by default to help you appreciate which of your marketing channels are actually bringing you the most goal conversions in your campaign. Isn’t that awesome?! In digital marketing, there are a few given models, like Last-Click or Last Interaction model where 100% of the credit for the conversion is attributed to the last marketing activity that led to the conversion. Another example of a model is the Time Decay version, where credit is shared in a progressive way, where more of the credit is given to those marketing channels that the visitor encounters later on into his journey to conversion.

So with Attribution Modeling, I quickly found an expression for my unique kind of football analysis. The fact that my analyses are NOT ABSOLUTE in nature makes me feel very much at home with Attribution. The constant learning and testing of Marketing Attribution models in Google Analytics can greatly sharpen your ability to improve future marketing efforts; which makes me very glad that I have always rebelled against football commentators’ and general opinions, but have always learned to chart my own course in assigning credit in football teams’ victories or losses.
Let me demonstrate:

LAST CLICK - The Barcelona Example:

I am not a Barcelona fan.
In fact I actually hated the club immensely from the early 90s right up until Pep Guardiola took over and turned them into the second most glorious footballing spectacle that I ever saw in my life. The most glorious being the Real Madrid era from 1995 to 2002.
Yes. I am a Real Madrid supporter; but I guess that other Madridistas might refuse to believe that, on account of what I just said about their rivals, FC Barcelona. But in such an event, all I can say to them is “grow up, and enjoy football as a game!” Hehehe.

Anyway, now I am going to demonstrate how paying close attention to Sergio Busquets, Xavi, Dani Alvez, Pedro and Lionel Messi, has coincidentally helped to greatly improve my understanding of Attribution in Google Analytics.
Now If you’re a fan, you’ve probably noticed that I left out Iniesta from this Barcelona selection. Well that is because this article is NOT exactly about football, but more about Marketing Attribution. And I have actually observed that Iniesta’s contribution to the Barcelona team’s overall attacking potency is much more of a diversion to the opposition’s defensive formation. And that doesn’t adequately help me in analyzing website macro and micro conversions.
So for the sake of this model, let us assume that Messi, Busquets, Xavi, Alvez and Pedro are 5 variables of marketing channels in an Attribution model. They could represent, if you like, PPC, YouTube, Banner, Email, and let’s add some shiny new toy like TourDash.

At FC Barcelona, Leo Messi is the guy who scores the vast majority of their goals. But what really makes this individual special is that he is also the one who provides a reasonable majority of their goals (at least over the past 4 seasons). But what might yet surprise most of you, is that I have also carefully observed that this same Messi is often also involved with providing the assist to the ASSISTS…what I am now call the 3rd Ball To Goal.

Meanwhile, this doesn’t automatically mean that without Messi, Barca won’t score goals, because football is not that simplistic. What it does mean though, is that those who have everything to gain from the team’s overall success (the coach, the owner of the team, and the fans)are forced to reckon with Messi, and will not likely be willing to gamble on leaving him out of the team for important matches.
Now, in an extreme case, if Messi’s contribution is overemphasized, and his team mates’ are assumed to be redundant; such a situation could illustrate the Last Interaction Model in Marketing Attribution.

Many business & website owners are fond of taking the reports of their web analysts superficially, assuming that the Last Interaction of each customer is the sole reason for his/her purchase or goal completion. Thus they are assuming “Messi” does it all, without digging deeper to find out how much of a role the other players are playing to assist him. The problem is that these clients may therefore increase their focus on that particular advertising channel; and reduce, or completely freeze off any efforts on improving other channels. Yet such a scenario could be very harmful.
So thanks to FC Barcelona, I came up with a custom model of Attribution. Which is merely a more conservative and perhaps more practical variation of the Last Interaction model.

From the diagram labeled Figure 1 above, you can see how I have reduced the over dependence on the Last Interaction, by assuming that other marketing channels may have played a more important part in the decision process of the customer. It is important to note also that the labeling of each color bar with a designation of the 5 various channels

Once you’ve done your goal setting in Google Analytics (GA) and all of your online marketing campaigns (and even your traditional marketing campaigns like TV, billboard & magazine ads) are being properly tracked; the next thing is to begin to analyze and do some tests with the new assumptions that you discover from your attribution models.

LINEAR ATTRIBUTION- Falcao Left Atletico Madrid:

Now any true football fan today as heard of a Colombian footballer by the name of Radamel Falcao, but those of you pure digital marketing nerds may simply just take my word on anything I say, for the sake of the argument okay? Hehehe!

So the funny thing about Falcao is that he was such a great striker that only a few people noticed how awesome his supporting cast was. The likes of Arda Turan, Diego Costa, Adrian and Koke made Falcao’s job spearheading the attack much easier that it looked. 

But now that the Colombian striker has left the team, they seem to have become even stronger.

Attribution helps you to understand that the visitor of your site that converted (i.e. took an action that you have indicated in GA as a goal) may not actually be experiencing your website’s marketing tactics for the first time. It doesn’t always mean that because a customer came into your site and purchased a camera because you sent an e-mail the day before, that it is automatically the sole reason why he took that action.

There is need to discover if there were probably several assists leading to that email. The same customer could have at various points taken a mental note to visit your website after seeing several of your other marketing campaigns like your blog, your Pay Per Click advert on Facebook, and maybe even your banners on affiliate sites.

Now, the Liner Model of Attribution is designed to give equal amount of credit to each of the marketing channels of your campaigns. Digital marketers use this when they assume that every member of the “team” is of equal importance. Ofcourse, this may also be a poor assumption that could lead to wastage of marketing spending on some channels, but the idea is to experiment until you find what works best for your campaigns based on understanding how your customers interact with your channels.


I hope this was at least slightly enlightening to you, and you’ve gotten the idea that Google Analytics is a great tool that you might have been overlooking in your marketing efforts.

Starting early enough is what really matters. Then we must patiently continue growing in understanding, in know-how and in experience. I have shared a few other articles in digital marketing that you might like to look at.
If you would like to pick more of my brain on other digital marketing topics, you may reach me on my Google+ profile and add me to your circles. You may also join my Digital Marketing Community

Alternatively, you might wish to see how opinionated I am about Football/Soccer on my G+ Alternate Page, True Football Analyst

I am Rotimi Orims. 
I am passionate about Exploring the Web, about Learning, about Digital Marketing, about Football, about Movies and about Music. 
I know, those are a lot of passions!