mmm-homepage

One of the first SEO guys I worked for instilled the virtue “Never build a website based on what you like”, which seems somewhat counter-intuitive, but you’ll never believe the the kind of stuff your potential clients really like. It often takes a lot of time, a lot of testing, and a lot of asking friends and family “whatcha think about this?”.  Today, most SEO’s will tell you with a techy tone “That’s what A/B testing is for.  Divide and direct your visitors to two different versions of your site, each one with its own slew of subtle changes, and whichever one performs better is obviously the better choice.”

A/B testing is truly one of the most major examples of “easier said than done” in the digital marketing world.  E-commerce sites benefit the most from this by simply taking a look at what products were clicked, which “buy now” buttons received actions, and what sold most based on color schemes, positioning, call to actions and much more.  What about the service industry, like the majority of our clients?  It is hard to always gauge the calls they receive, which is often the overwhelming majority as the form of contact.

And then what about traffic.  That’s the big one.  Many small to medium sized businesses don’t get a staggering amount of traffic, so a rational A/B test could possibly take weeks, if not months.  Who wants to wait around that long?  What am I really leading up to?

Introducing: an alternative to A/B testing through goals, heatmaps, and analytical traffic flows.  Added bonus: direct feedback from real people!

We are using Amazon Mechanical Turk (https://www.mturk.com/mturk/welcome) as our go to for putting these tests together (currently researching for alternatives now too).

Essentially, this is your opportunity to ask random people what they think, what they would choose and prefer.  In this situation, Chase and I were unsure about what people besides ourselves would think about a home page with a larger or smaller image slider.  First of all, we’re letting others in an unbiased environment decide based on the question “What would you prefer if you were looking for the services of a marketing company”.  Second, we are getting an idea within only a few days instead of waiting a few weeks.  This enables us to tweak much quicker and make adjustments while saving a ton of time.  Imagine you were selling a hot new product and you wanted the CTR of the “Free Trial” button on the landing page to get to around 40%.  It was currently at 15%.  After every A/B test, the CTR looked like this:

Tweak 1: 15% to 24%
Tweak 2: 24% to 30%
Tweak 3: 30% to 33%
and so on….

How important would it be to you to get the immediate feedback you need to make educated changes, instead of analyzing and surmising the behavior after pouring over a week’s worth of data?
I’d say pretty important.

mturk-results

These HIT’s (Human Intelligence Tests) can go as cheap as 5-10 cents, so you could have a ton of great data for as little as 50-100 dollars.  I probably sound like I’m getting kickback for promoting the product, but honestly the possibilities are extensive and exciting for nearly any business.  Some other ideas for this test….

  • Ask people to choose between 3 CTAs and decide which is the strongest.
  • Images of contact forms, buttons, banners and find which is the most appealing.
  • Development of a new product or service.
  • Making adjustments to the current ads/banners you use for a retargeting campaign or placement on a website.

Good ol’ marketing research has never been so accessible.  If you want to learn more about how your company can be utilizing HITs for your own research, or are interested in us helping you set up one of these campaigns, don’t hesitate to reach out to us to learn more!