The Growth of Experimentation chart

6 of one, half-a-dozen of the other. Multivariate testing proves users prefer “6 of one”.

posted in: process | 0

Title: The Surprising Power of Online Experiments
Context: If at first your design doesn’t succeed, multivariate test it against a control option.
Synopsis: We’ll never admit it in polite company, but the truth is that a fair amount of the work designers do is guesswork. Of course, it’s based on years of experience, thousands of hours of practice, and innate talent for problem solving, but decisions are still made based on intuition, assumptions, and often a fair bit of hopeful optimism. And that’s ok. Interaction design is still as much art as science. Failure is an option, as long as the opportunity and desire to correct errors exists. Of course any ability to shorten—or even, dare I say: eliminate!—the gap between the commission of an error and its correction would be the magic wand all decision-making designers wish they possessed. Well, in fact, we already do. Through the power of data, analysis, research, and iteration, no longer do we need be defined by our mistakes but can now rather be celebrated for our resiliency. Being wrong isn’t the problem. Not making things right, is.
Best Bit: “Controlled experiments can transform decision-making into a scientific, evidence-driven process—rather than an intuitive reaction.”

via hbr.org

Leave a Reply

Your email address will not be published. Required fields are marked *