​Don’t be a tweaker--seriously

Today I’m going to pick on what I call the “tweakers” of behavioral science.

Another name for people could be the “cookie monsters”—those that inelegantly copy tricks that they’ve read about in papers or books. They’re masters of cookie-cutter tactics.

Their M.O. is something like this:

They’re given an email-optimization task. “We want you to increase the conversion rate of our weekly deal email”.

Instead of thinking deeply about the problem at hand, and how to create a meaningful and compelling piece of content that speaks to the true needs and desires of the customer base, the tweaker reaches into his bag of tricks.

“So you want to increase conversion, eh? Try out some loss aversion, or some social proof.”

They slap an “URGENT: OPEN NOW” or a “Most customers are buying these things” in the subject line, hoping that these tried and true principles (from Cialini’s “Influence”) will, once again, win the day.

The only problem is that these tactics are usually quite awkward, and usually result in just as many costs as benefits.

In the case of the “urgent” messaging, you better have something that’s *actually* urgent in the body of the email. If your customers open up the email and find out that it’s just a spammy message, “these hot deals are about the disappear!”, not only will they not buy… but they’ll be less likely to open your emails (or trust your company) in the future.

That type of messaging should only be used for things like billing/credit card failures, expiring credit, and so on. And, even then, it’s a bit of a stretch.

Don’t even get me started with social proof messaging… I’ve seen so many companies try to bullsh*t their way to social proof, it’s truly astounding. Social proof only works when something is *actually* the overwhelming social norm. If only 10% of your customer base is doing something, you can’t really use social proof (unless you torture the data so thoroughly it becomes absurd).

But “tweakers” will torture the data if it allows them to use one of their handy-dandy tactics. They’ll find out that 60% of the customers that signed up on the week of July 7th(the week you signed up) bought a weekly deal item last year, and so they feel justified saying “Most of our customers like you purchase a weekly deal”… even though only 10% of customers actually do such a thing.

Not only do I find that thing kind of sketchy, but it also generally results in crappy content. If social proof is going to be the centerpiece of your email, how interesting can it get? “By the way, did we mention that this is what most people do!?”

I find that these behavior tweaks often shut down any sense of humanity or creativity in the individuals wielding them—but tweakers will do anything they can to get their quick “fix”.

In my mind, the right thing to do in these situations is to look at the problem as a human. Come up with as many good ideas as possible. Think: what would I like to get in my inbox? What would really speak to me? What would be fun? What would be compelling?

Try and get into the heads of the people you’re trying to influence, and then put *care* into creating some messages that you think they’d like. If you’re trained in behavioral science, it’s natural that many of these ideas will contain behavioral-hacks or concepts… but they won’t be the center of the show. They’ll come out naturally, in service of the greater message you’re trying to get across.

If you’ve been thinking about how people *actually* think, the science of human perception and behavior, for some time, it will express itself in everything. The types of ideas that I, a behavior nerd, will have will be different from those of a sociologist or art history major.

Many of their ideas, in any given brainstorm, will be more elegant and creative than mine—that’s just the nature of the creative process. But my ideas will be more likely than theirs, in the long run, to actually produce behavior-change results. Why? Because I’m operating with a more realistic and truthful model of human nature—one that I’ve constructed after well over a decade of hard study and work experience. My operating system in my noggin will put out ideas that are better, on average. But that doesn’t mean I’ll win every time.

Yes, I can consciously use a bunch of behavioral-science ideas to further optimize and refine the ideas that I (or others) come up with… but jamming these hacks and tactics into elegant, fully formed concepts can, as I said before, often do more harm than good. And basing entire strategies around a single tactic or two can result in abominations. In my mind, the best role of behavioral science is as an operating system… and, like an operating system, it does a lot of its work in the background. It provides us with a worldview that allows us to make much more accurate predictions, and come up with new creations, than our colleagues and opponents.

It's like a quantum physicist (behavioral scientist) going up against a Newtonian (non-behavioral scientist)--yes, the Newtonian is going to be able to make pretty good predictions about a lot of things... but there are entire domains of existence where the Newtonian is going to get trounced. The Newtonian just doesn't have the right tools to explain what's happening at the subatomic level, just as the layperson doesn't have the tools to explain how thinking, memory, and a host of other fundamental cognitive processes actually function. While this may seem like a crazy comparison (and I admit it's a bit of a stretch), I do believe that those rigorously trained in the psychological sciences have a huge advantage over those with their own cobbled-together folk psychologies; and that this will result in more and more of the upper-brass coming from behavioral backgrounds.

And this can't come quickly enough. Because, until the day that the top brass adopts this new operating system, the tweakers will be able to roam free; inappropriately applying their tactics to projects and products across the land.

Jason Hreha