The causality fallacy

I was never that good at statistical analysis in school. In fact, it was the very last exam I did before graduating (and it was a first year class, so you get where I'm going with this). I also never raved that much about the scientific method applied to social sciences, even if my grades were good at that class. It's fair to say I now feel the pain of never having truly embraced these two subjects (I'm working hard to restore balance in that matter), because they are part of the reason we've been doing our job as marketers and strategists and community managers all wrong.

Take a minute to read this Business Insider piece about how Samsung reportedly generated $129 million in smartphone sales "due entirely to Facebook". It sure seems like a great achivement both for Samsung and Facebook, specially since the former has been in a quite interesting duel with Apple (some might say Samsung is winning) and the later has been taking hits like this one or even this from yours truly regarding latest moves and less than optimal results, according to Wall Street standards at least.

Except we tend to oversimplify complex issues to a single perspective analysis, which ultimately leads to poor justification to why things happen the way they do. Typically, if A then B and if B then C, then the existence of A will necessarily lead to C. Given some context, though, we might be wrong in thinking like this.

Going back to Samsung and Facebook. Some highlights from the article:

Facebook Global Marketing Solutions VP Carolyn Everson just revealed at Business Insider's IGNITION conference that Samsung spent $10 million over three weeks — during Apple's iPhone 5 launch — for its Galaxy S III release, and made $129 million in sales, due entirely to Facebook.


Samsung specifically picked Facebook for the launch, Everson said. It makes sense, Facebook reaches three times the Super Bowl audience every day.

Ok. Now let's go back a few steps and ask ourselves some serious questions (which a serious journalist or editor would also ask before writing this article, something Business Insider doesn't seem to have done): Were there other media investments? Which media? How did they split the marketing budget across all touchpoints? What percentage of sales is attributed to other media besides Facebook?

If you've been paying attention to Samsung's product launch strategy, there were video ads, print ads, PR efforts and lots of media coverage across tech blogs and various news outlets. Bloggers covered the product and reviews overall have been good. So, basically it was a successful product launch due to a series of successful factors. Key point: "a series of successful factors". Which curiously were not mentioned in said article for some reason (I will not go into this particular issue right now). "Due entirely to Facebook", they insist.

My problem here isn't necessarily the reporter or editor or website which released the article, and it's not Samsung either. It's the way we as professionals are so eager to prove some point that we quickly sip any information that will help us look good in an argument about the power of social media or media planning or whatever is the discussion du jour, even if the information is obviously incomplete.

The biggest obstacle strategists, marketers, consultants and analysts now face isn't lack of tools or insights or consumer touchpoints. It's poor statistical analysis and complete disregard for the scientific method.

Poor statitical analysis because we easily mix correlation (including Facebook on an integrated launch strategy is correlated with the fact that there was a high volume in sales due to that strategy) with causality (since Facebook was included in the strategy, Facebook was responsible for a high volume in sales).

Disregard for the scientific method because we easily attach the outcome of a campaign to a single factor or to a small number of factors, ignoring other variables which may have contributed to that same result. "Integrated" is more than a PR-able buzzword you can place in your press-release or the mere aggregation of different channels in your strategy; it's the very way we conceptualize, plan, execute and measure (I repeat, and measure) that same strategy.

I call this the causality fallacy, and it's a real issue that affects many digital marketing professionals. Perhaps more than figuring out which are the hippest platforms for 2013, which trends should we blindly follow or how to create the next big meme, we should look at ourselves and see if our way of thinking, before our strategies, is truly integrated.