Why Don’t We Test ALL the Things!? A/B Testing and Beyond

Joan harris problem

“I’m not a solution to your problem. I’m another problem.”

In the time of the Mad Men – the 1960’s – marketers were faced with a problem: not knowing how to really tell what marketing worked and what didn’t.  Often, even up to the present day, marketers make decisions that sometimes don’t have the effect they want, sometimes those decisions do the opposite of intended and create another problem.

BUT it’s getting better.  And A/B testing is a part of that.

A/B Testing is essential.  It takes most of the guess work out of optimizing your marketing materials online.


The image above shows what it is.  You take 2 variations of a website, with one small change and split your audience so that they evenly and randomly see each page.  Then you see which version tested better (statistically significantly better).

But if it is so useful, why are we not doing more?  Why can’t we find ways to test everything?


I was reading an article on Forbes recently about a content marketing study which suggests that most content marketing doesn’t work.

As someone who intuitively sees the value of  content marketing and would by far prefer that over traditional marketing, I had to remove my bias.

The study was conducted by InboundWritier for a year and looked at dozens of websites.  They found that “Only 10 to 20 percent of a company’s website content drives 90 percent of its web traffic, and only half a percent of a website’s content drives more than 50 percent of its traffic.”

So basically, they’re saying that only a small amount of content actually does anything important.  I think that the other 80 to 90 percent of content has its purpose – to give a company legitimacy or authority – but still, it would be great to know that it was also doing work for you, right?

InboundWriter has tools using predictive analytics to “score” content before it is published, allowing marketers to edit as needed.

How much better would you feel about hitting the publish button if you had that tool in your tool-belt?

In the end – I think that the tools marketers have for testing effectiveness will be growing in the years to come.   Hopefully this will mean that content will become even higher in quality.  My only fear is that sometimes giving the masses what they want is not a good thing.  After all, 3.2 million people watch Honey Boo Boo, that has to say something.


Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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