The truth is that no one knows what will work best for your marketing.
I don’t, you don’t, and even your customer’s don’t.
To get around this, you have to create a hypothesis, run tests, and analyze the available data.
You make little bets, and try to determine what will work, and what won’t.
During these tests, getting feedback is important. It helps you gather insight from real users. And it will open your eyes to areas you may have overlooked.
But it can only get you so far.
Image courtesy of Fred Seibert
How Sexually Stimulated Students Choose
People don’t always make rational decisions. This point was highlighted in the brilliant book Predictably Irrational.
The book’s author and a fellow researcher wanted to find out how being sexually aroused would affect the decision making of college students.
They asked Berkeley students a series of questions both in a normal state, and in a state of sexual arousal.
The results were shocking.
Here are two of the more interesting results:
1. Could you enjoy having sex with someone you hated?
When sexually aroused, people were 22% more likely to have sex with someone they hated.
2. Would you use a condom even if you were afraid that a woman might change her mind while you went to get it?
When sexually aroused, people were 26% less likely to use protection if the other person might ultimately change their mind.
The findings are clear, and pretty significant. The truth is that our emotional state alters how we decide.
It doesn’t matter if we’re angry, stressed, overjoyed, surprised, or… aroused.
No matter what people tell you, they may actually do the opposite depending on their “current emotional state”.
Forget Feedback, and Focus on User Behavior
Verbal feedback is helpful… to a point.
To really get the whole picture, you need to focus on how people actually behave.
Every time someone does something, or doesn’t do something, it’s a piece of evidence. Your job is to gather these clues and connect the dots.
Observe how people act, and adjust accordingly.
For example, you can use a tool like CrazyEgg to see where people click on your website and how far they scroll down a page. That will tell you if people are actually reading the entire page, and whether you should make it more compact.
You can also use basic website metrics like Time on Site to infer meaning. If people spend less than one minute on your site, and they aren’t converting, then you have problems. What can you do to make people stick around longer? Remember that you have to turn their attention into interest before they’ll sign up for anything.
The Classic Website Redesign Example
One of the best examples of this problem is when you do a website redesign.
If you’ve ever been involved with one, then you’ll recognize the familiar problem. There’s always too many cooks in the kitchen.
Let’s say that the marketing team wants the call-to-action button green, while the President wants blue, and the customers voted red in the latest survey.
Who’s right? The marketing department, the President, or the customers?
None of the above.
The answer: Whichever color performs best.
You can run split-tests to find the real winner. If one of the colors converts at least 5% better than the others, then you have a winner.
And it’s time to pick a new variable to test.
You can’t really predict what will work best before you start. No one has the answers. And your customer’s verbal feedback in this case is useless.
The real key is to analyze user actions and behavior in the context of your goals.
You can use this insight to make important decisions, and you’ll have a much better chance of being right.