Good, sound reasons or reasons that sound good?

Campaign tracking for a test email I sent -- not the best results, but with this data, I was able to tweak it for better results on the next message.

I was talking to a young marketing maven the other day.  She’s a very smart young woman, with a wide-ranging set of skills, and she does great work for the advertising & PR firm where she is a rising star.  But as she attempted to explain to me the logic behind a marketing campaign she’d put together, I heard the voice of my 8th grade Latin & logics teacher saying, “Correlation does not imply causation – don’t make a logical error in your thinking.”

My young friend was talking about a multi-channel marketing campaign she was running, and trying to explain how she convinced her client to invest heavily in one particular part of the project – only to find out when the campaign ended that the ROI on that spending was actually quite low.  She had a good argument, told her story well, and had managed to placate the client and get her next project budget approved.

But as Laurence J. Peter wrote in The Peter Principle, “There’s a mighty big difference between good, sound reasons and reasons that sound good.”  You may have awful results even If you base your marketing spending on good, sound reasons — but if you base them on reasons that just sound good, no matter how good the results are, you can’t replicate them becuase you won’t understand the reasons why the tactic worked.

Business social media isn't about "conversations", it's about engagement. If you can't measure it with tools like the Socialware Voices dashboard, why are you spending money on it?

That was my friend’s problem.  She had no clue why the marketing campaign didn’t work, and in looking at the plan she’d asked me to proofread for the client’s next big project, it was clear to a more experienced eye that there was no logical strategy behind the recommendations in it, either.  

Why?  Because she was committing the two most common logic errors that Laura Blix Jones warned us about in 8th grade. 

The first problem my friend had, post hoc ergo propter hoc, Latin for “after this, therefore because of this”, links two factors with the assumption that because one happened after the other, the first one caused the second.  For example, “I ate ice cream.  Fifteen minutes later, I fell and broke my arm.  Eating ice cream caused me to break my arm.”  Stated that way, the logical fallacy is obvious – and no one past kindergarten age would generally make that link. 

But people (especially politicians and those trying to sell me something I don’t want or need) frequently use logical fallacies to confuse and convince. Sometimes it’s because they weren’t thinking clearly in the first place, but I suspect that some of them know very well that there isn’t any logic behind what they’re saying — they’re purposefully offering an argument that sounds good instead of a making sound argument.

Here’s an example.  The HPV vaccine prevents both an STD and cervical cancer.  One candidate said that she had heard the shot wasn’t safe — “one girl who took it became mentally retarded.”  It’s a ridiculous logical fallacy with absolutely no basis in fact – but unfortunately, when a presidential candidate says it from the bully pulpit of the campaign trail, she puts the health of millions of girls and young women at risk by frightening parents into not vaccinating their daughters.

The second problem that was clear in the marketing plan my young friend presented is cum hoc ergo propter hocThat’s Latin for “with this, therefore because of this.” This kind of reasoning says that the chronological ordering of a correlation is insignificant and that two events that occur together have a cause-and-effect relationship.  In her plan, she recommended a large boost in spending on the company’s blog because the blog got 100,000 visitors in the same week that their new product made its debut.

Well, duh, right?  Launch an exciting new product, and you get more visitors to your blog.   Trouble is, when I checked, the company blog made no mention of the new product that week – the first blog post about the new product appeared on the Monday AFTER the product announcement. (Waiting a week to mention the company’s new product on the company blog is a different kind of problem — one of scheduling and planning, not logic.) There was no link to the company’s blog in the press release or ads for the new product which appeared the week that the blog experienced a large surge in traffic.  Could people have used organic search (entering the company name and the word “blog”, for instance) in an attempt to see whether there were more details about the new product on the blog? Sure.  But did they?  Not according to Google analytics.

The Google analytics report for the week with the traffic spike showed that an article that appeared in ZDNet Asia, which linked back to the blog, was responsible for over half of the traffic on the blog that week.  In other words, there was no cause and effect relationship between the new product announcement and the spike in blog traffic.  So there was no link between the popular new product and the spike in blog traffic — and no link between the spike in blog traffic and the popularity of the new product. 

I tried to explain the logical fallacy to my young friend. She was too polite to say it, but she was probably thinking, “What does a bunch of Latin have to do with social media, email, PPC, blogging, and mobile marketing?” 

Nothing, of course – but how you think when you plan your campaigns has everything to do with the results you can expect.  If you aren’t thinking clearly, and linking the right cause and effects backed up by solid analytics, then you won’t be spending money on the right things. If she wanted to spend money on the activity that caused a spike in her blog traffic, she should increase spending on PR — not the blog, and not her product launches.  At least, that’s what the data tells us.

Post hoc is a particularly tempting error because temporal sequence appears integral to causality. The fallacy lies in coming to a conclusion based solely on the order of events, rather than taking into account other factors that might rule out the connection.  When I first finished college, we didn’t have access to real-time measurement and analytics tools.  It took months – and a lot of hard work and money – to measure and report accurately on a campaign.

Avinash Kaushik is an analyst at Google; this is his Google Analytics Dashboard, which was made available by Google's PR department.

Now, digital campaign measurement is built-in.  My email marketing management tool (I use the Distribion DMP), my social media campaigns (BufferApp) to schedule four separate Twitter accounts at the optimum “tweet time” for each group of followers, and I like Socialware for overall social CRM), and my lead generation efforts (the Distribion DMP and are all measured and reported in real time. PR and social sharing via MyPRGenie come with real-time analytics, too. 

So do all the other applications I use to manage my marketing, from customer preference tools from Gryphon Networks to data services from Epsilon Data to specialized services like SEO and PPC which I purchase through Vertical Nerve.

So there’s no real excuse for logical fallacies on cause and effect.  The data is there, and all I have to do is review it without pre-conceived ideas about what it means.

About debmcalister

I'm a Dallas-based marketing consultant and writer, who specializes in helping start-up technology companies grow. I write (books, articles, and blogs) about marketing, technology, and social media. This blog is about all of those -- and the funny ways in which they interesect with everyday life. It's also the place where I publish general articles on topics that interest me -- including commentary about the acting and film communities, since I have both a son and grandson who are performers.
This entry was posted in Uncategorized and tagged , , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.