Logo - Full (Color)
Skip to content

Episode 3: 4 Things to Start A/B Testing Right Now

It can be tempting to just guess which type of content will entice visitors most and eventually convert into leads. But basing your marketing decisions off a hunch is risky business.

A/B testing helps you compare two versions of something against each other to determine which one performs better. Anni lets you in on the three things you should start A/B testing right now.


 

Listen Now

Episode Transcription

Hi. I’m Anni Kim from HubSpot, and this is Skill Up, the show where you’ll learn how to take your sales, marketing, and service skills to the next level.

If you’re a marketer, you’re creating content. It’s basically the core principle of your tribe. Sorry, Mando. This is the way.

When marketers create content, they have a lot to think about.

It can be tempting to just guess which type of content will entice visitors most and eventually convert into leads. 

But basing your marketing decisions off a hunch, Tom Cruise, is Risky. Business. [nailed it]

Rather than just relying on assumptions, it’s a much better idea to run an experiment on your content. 

Look, different audiences behave differently. Something that works for one group, may not work for another. That’s why marketers run what’s called. A/B testing.

A/B testing is a method of comparing two versions of something against each other to determine which one performs better. 

Sometimes, folks call it split testing, but you and me, we’ll just refer to it as A/B testing. 

This type of marketing experiment splits your audience 50/50 between two variations of a page. Then, you decide on a winner -- which, sure, is a real loose definition of ‘choose.’ You’re picking the one that drives the most traffic, clicks, or whatever marker of success you’re looking at.

So I hope you invested in that white labcoat you’ve been eye-ing all these years. Because with A/B testing, you can experiment with changing different variables on your pages, and see what resonates most with your audience.

Companies like Microsoft, Amazon, Google, and Facebook each conduct more than 10,000 online controlled experiments each year. And these kinds of experiments can have incredible results. 

Time jump real quick. Back to 2012. The Obama campaign has raised an additional $75 million dollars from marketing decisions made by running A/B tests on their website. 

Running experiments on your content can be a great way to learn how to effectively drive traffic to your website and generate more leads from the visits you’re getting. 

Even small tweaks to your pages can affect the number of leads your company attracts and converts. Use the insight you gather from your experiments to inform your marketing decisions moving forward and drastically level up your content. Ah, it’s like it’s the tagline for our little show here.

When experimenting with your content, the first thing you need to decide is, “What am I going to test on this page?” In any test, there’s a control and a variable: The element that stays the same is known as the “control” and the element that you change, in the hopes of improving your results, is known as the “variable.” 

The possibilities? Endless. Countless. Make as many changes to your heart’s desire to test page effectiveness. You’ll also come to find out, it can quickly become completely overwhelming. So it’s important to remember: not all variables are created equal -- some are worth more of your time than others.

So let’s focus on a few essential elements that tend to bring the biggest returns: CTAs, copy, layout, and media. Let’s take each with a bit more detail.

For your CTAs, keep these three principles in mind: easy to spot, action-oriented, and easy to understand. Try making small changes to your CTAs to test how effective they are in each of those areas. For example, make your CTA a color that contrasts with the main color scheme of your page. Or, use action verbs like “Download” or “Register” to compel visitors to click your CTA. 

Next up? Page copy. In the words of Nora Ephron, ‘Everything is Copy.’ So to be more specific, we’re talking web page copy. These words are critical to capturing a visitor’s attention and making them want to learn more. 

Your copy should be clear, compelling, helpful, and organized. Make the value proposition on your page as clear as possible, and use action words to excite people and encourage them to read on. Try using headers, paragraph breaks, and lists to break up blocks of text and make the content easier to consume. 

Another element to A/B test is your page layout. Clearly laying out the content of your page can help you highlight the value of your offer and increase conversion rates. 

The layout of your page should be value-focused, logical, and uncluttered. Stick to this principle: Simple is always better. People visiting your website are busy. They’re not looking to sift through info to find what they’re looking for. 

Make it easy for people to get answers to their questions and solutions to their problems by highlighting the value of your offer right away. Use headers, sub-headers, and easy-to-scan bullet points to explain your offer clearly and concisely. 

Get to the point quickly, and stick to it. 

Finally, you can A/B test the media on your page. Adding visual media like images and videos is a great way to boost engagement. However, you want to make sure you’re adding the right media. High quality, authentic, and optimized media is best. 

So make sure that all your media is high quality and conveys a professional image. Keep things authentic, with real life images of you and your team at work. That means no stock photos. Never. Nunca. And always, always, test your media to ensure it’s optimized for mobile devices. I mean, it’s 2020 at this point. If not already optimizing for mobile….

And that does it. Remember, if you try to test too many different elements at once, your results might be inconclusive. So start with one of those A/B tests, and build from there.

So to summarize...

A/B testing is a method of comparing two versions of something against each other to determine which one performs better. Sometimes called split testing, this marketing experiment splits your audience 50/50 between two variations of a page. Then, you decide on a winner. 

With the right software, you can use AI to run adaptive testing. Sounds fancy, but the theory remains the same. Adaptive testing is a type of experiment that uses machine learning to run through multiple variations of a page and automatically distribute traffic to the best one.

Machines. Always learning.

Next week, we leg it north of the wall to tackle giants, because we’re closing big, complex deals on a small target of accounts. Better known, as Account-based Marketing.

See you there!