Testing is the hallmark of any good online marketer (and of any marketer, really).  How many times have you heard us espouse the benefits of testing at the Internet Marketing Monitor?  It can't be stressed enough.  Successful business are constantly testing and trying to improve.

But as Bryan Eisenberg points out, testing is a science.  And in "We Tried That Already", Eisenberg looks at some common testing mistakes that are often made by those who don't fully understand the science.

These things are easy enough to do.  I know a little about chemistry.  Actually, I probably know more than a little about chemistry.  I was a professional student for years and had more than your introductory exposure to the subject.  But chemistry is a science that requires more than "a little" knowledge to master.  I'd never tell a chemist that I'd tried something already without finding out if what I'd done was really done right.

Testing in the marketing world is no different.  Maybe you tried something in-house that didn't work well.  Maybe you tried something a year ago that didn't work well.  Or maybe you've avoided making changes to something because you think that you're already doing it as well as you can.

These are all topics that Eisenberg looks at in his article. But by far my favorite of the topics that he covers is "Over-Conforming to Best Practices":

Best practices should always be looked at with some degree of detachment. If you're operating an average, run-of-the-mill business with average, run-of-the-mill customers then best practices might turn out to be the best practices for you.  But more often than not, what's best for other companies isn't always what's best for yours.  You can learn from them… you can use them as a starting point.  But don't just assume that a problem exists if your company doesn't operate within the guise of the so-called "best practices".

Eisenberg concludes the article by stressing that the "why" of an unsuccessful test is much more important than the "what".  Every test… even an unsuccessful one… is a learning opportunity.  Contrary to a lot of other folks, I'm actually a big fan of mistakes.  A mistake is by far the best way to learn something.  And as long as we look at mistakes and "unsuccesses" from a learning standpoint, they're never a bad thing.

It's easy to identify what didn't work.  It's a little harder to figure out why it didn't work.  But that "why" is where the learning experience from the test comes into play.  If you don't figure out why your modification didn't work, you've effectively rendered the test a complete waste.  It's not always as obvious as the what… but the why is much more important.

Be sure to read Eisenberg's entire article.  I've just hit on the high points here and there's a lot more to learn in the post.