How To A/B On Shopify

Hey! Welcome back to Brand Building With Nichola. This week I’m diving into A/B testing on Shopify and the exact process my team and I go through to optimize our stores.

As a Shopify CRO agency, it’s crucial to have a standardized process to run A/B tests so we can ensure a smooth setup, execution, and most importantly clean data.

But what is conversion rate optimization (CRO) and why is it important?

CRO is a lot more than just improving your conversion rate. At the micro level (ex. A specific landing page) it can be as simple as that, but at the macro level, performance CRO focuses on global website performance to optimize for your specific KPIs.

For the vast majority of eCommerce brands, this is profit!

CRO is important for your brand because it has a direct impact on your ability to scale your other marketing efforts such as Meta ads, TikTok Ads, email marketing, PPC, and more!

To put it simply, the better your site (post click experience) is, the better your other marketing channels will perform.

For a hypothetical example, if Meta drives 1,000 people to your site and only 10 people convert, this is a 1% conversion rate. But if Meta drives that same 1,000 people to your site and 20 of them convert, your conversion rate is 2%, which with all other metrics being equal, doubles your ROAS and cuts your CAC in half.

Keep reading to learn how you can use our 7-step Shopify CRO process to start seeing massive improvements in your store’s performance while lowering your CAC and increasing your ROAS.

Identify the phase of the buying journey to improve

The first step is to identify what needs to be improved on your site. An easy starting point is to look at where you are sending most of your traffic. For most brands, this will be your product page, collection page, or a landing page.

Then by looking at your analytic tools such as Shopify analytics, Google Analytics, surveys, and heatmap tools, you can start to identify what to improve.

Is there a pinch point in scrolling? Are people not finding all of the information they need?

PRO TIP: Remember, your store needs to answer the questions people have in order to make an educated buying decision.

Identify the key metric to improve.

What do you want to improve? For most brands, the key metrics to improve are conversion rate (CVR) , average order value (AOV) , and my favorite, revenue per visitor (RPV).

Revenue per visitor is a composite metric between conversion rate and average order value that give you a clear insight into overall website performance trends. To put it simply, if your RPV is increasing, your store is getting better.

Other metrics that people look at include email sign up, quiz completions, returning customer rates, items per order, and more!

Hypothesis the test to influence the metric

Now that you know what you want to improve, it’s time to come up with an improved design that you believe will have a positive impact on your metric.

For example, if you want to improve your add to cart rate, you could try some new copy or a more educational buy box to give your customers more information to make their buying decision.

Before you randomly update based on a gut feeling, it’s important to look at your data (customer surveys, reviews, FAQs) to improve your section based on real world feedback and actual custom use cases.

Design and develop the variant

Once you know how you’re going to improve your metrics, it’s time to design the store update then develop it. If you’re using base elements in your theme you can usually skip the design phase. 

However, oftentimes you’ll need to work with a custom developer to build more CRO specific elements for your particular brand.

Set up the test with Intelligems 

Once you have your variant ready for the real world, it’s time to set up your test. At 253 Media, we like to A/B test with Intelligems.

Instead of just trying to monitor your analytics in Shopify or Google and sending all traffic to your new page, Intelligems helps you track each metric individually in real time by sending half of your traffic to your original  (control), and half to the new variant. Then you can see in real time which option is performing better.

Plus you can track the metrics you care most about!

Monitor until statistically significant

One of the most common questions we get asked is “how long should I run my A/B test?”

Well… it depends!

To make sure you have clean data and your data is actually accurate. You need enough traffic to reach statistically significant results, which refers to a way to determine if the difference between the control and test versions of something is real or due to chance.

Implement the test if the hypothesis is correct.

Once you reach statistical significance, you’ll know if your hypothesis was correct.

Did your metric improve? If so, you can implement it!

But wait…

What is my target metric improved but another metric worsened? Then we look towards your profitability.

Thanks to Intelligems, we can see the bottom line profitability change to help us determine the true winner.

Now that you know how to A/B test like the pros, you can start making educated improvements to your store so you can get the most out of your marketing efforts.

Need some help setting up your A/B testing? Click here to book a call with our team.