Why Did Your A/B Test Fail?

marketing, digital marketing, blogging

 

A/B tests can exist to give you insights into a number of things about your marketing.

They can tell you whether your designs are grating. They can tell you if your offer is weak, if your language is uninteresting, or if your CTA doesn’t stand out.

You know you should be A/B testing, but doing it right can be tricky.

Let’s explore some reasons why you’re A/B test didn’t give you the insight you were looking for.

Your Offer is Weak

Your conversion opportunity exists for one reason: to get people to convert. If you aren’t offering something people want, they aren’t going to convert.

Human behaviors are motivated (at least in part) by the urge to gain something desirable, with a minimal amount of pain or effort.

It’s simple, really.

But figuring out what will get your target audience to convert isn’t so simple. When building your offer, ask yourself these questions:

  • Which pages on your website get the most traffic or have the lowest bounce rate?

  • Which products are your most popular?

  • Which offers can draw your potential customers deeper into the funnel and into more meaningful interactions with your brand?

While it’s true that your offer should be valuable to your customers and fit with the ways they’re already engaging with your brand, it should also offer an opportunity for more meaningful interaction.

A one-time discount for a product that’s not terribly popular? It won’t cost you much money, but it also might not be interesting or appealing to your customers.

And a product that isn’t terribly popular might not do much to keep your customers coming back after they’ve tried it.

Conspire to make an offer that will be irresistible to your potential customers and also beneficial to your company.

A/B tests are a great way to find out which offers your audience will respond to.

But you have to be willing to compare offers, stipulations, and time-frames, pitting them against one another in the court of your potential customers’ response.

If you only have one offer, and none of the versions of your landing page is converting, your offer might need some tweaking.

You’re Testing Too Many Things

In order to be successful, your test needs to be specific.

Say you’re A/B testing a landing page, and between version A and Version B, you’ve changed the language, the location of the CTA, the deadline of the offer and the wording of your headline.

When you changed this many things, your test isn’t likely to give you any valuable, specific feedback about your page.

Why?

  • When one version is so different from the other, you’re comparing apples and oranges. Sure, one might perform better, but with so many things different between the two, you can’t know what pushed the increase in clicks.

  • When you don’t know exactly what made one version better than the other, you can’t apply the data from your test to any of your future offers.

  • You’re spending time and money to make one version very different from the other. Save your pennies and your office hours by choosing one specific thing to change in your test.This will narrow your results and provide you with more insightful data.

When it comes to A/B testing, keep the variables limited.

You Don’t Have a Big Enough Audience

If you’re a small business or a start-up, with a brand new website and very little brand recognition, you probably aren’t generating enough traffic to run a substantive A/B test.

If orders or sales are your success metric, you need a substantial amount of traffic to generate enough usable data on your conversion pages.

Say your site receives around 1,000 visitors a month. Your landing pages probably receive far fewer visitors.  

This means your sample size for an A/B test is very small. It also means running an A/B test on a particular landing page could take weeks or even months, even then yielding no real results.

This test duration calculator can help you figure out how long your test will take to complete, based on a number of key components.

There are a few ways to make an A/B test more achievable for small businesses or websites with little traffic:

Change your success metric.

  • Engagements, like clicks to the next page in your conversion series, can serve as more easily measured (and obtainable) stats for your test.

  • For A/B testing on your homepage, clicks to any page are easily measurable for an A/B test.

Spend money to drive more traffic to the pages you want to test.

  • Drop some money behind a Facebook ad campaign or boosted posts to help increase reach and drive engagement with a targeted audience

  • Put some money behind Google AdWords to drive more traffic to your landing pages.

Like all important marketing decisions, your landing pages, website and offers need to be tested.

Finding out what your potential customers respond to and want is among the most valuable information your company can have.

A/B tests can help you get there. Start collecting data today.

Still have questions? Plan Left can help.