7 Top Ways to Use A/B Testing with Magento
As his business grew, he opened 15 more locations which later became a part of Macy’s. Many now consider Wanamaker a pioneer of marketing.
Here’s an interesting quote from him about advertising at the time, “Half the money I spend on advertising is wasted; the trouble is I don't know which half.”
Nowadays, thanks to the readily-available tools for marketing experimentation, we don’t have that problem. Through the likes of A/B testing, we can measure the effect of marketing and conversion rate optimization methods precisely. A/B testing can help you test design elements, copy and more to improve user experience and make more sales.
Here are the best ways to use it for the Magento e-commerce platform:
1. Perform Research to Inform Hypotheses
To carry out A/B tests that are actually going to help grow your e-commerce business, you need to perform research and analysis first. That way you can come up with solid, data-driven hypotheses to test. You might think changing a design element will drive more sales.
For instance, there have been tons of case studies that supposedly prove one color CTA button converts better than another.
But you shouldn’t be lured into creating arbitrary tests based on the research of others. What works (or doesn’t work) for your site and your customers is all that matters.
Through research, you can prioritize your A/B test hypotheses and ultimately improve areas that make a real difference to your e-commerce site.
So, let’s take a look at what kinds of research will best inform your hypotheses.
Analytics
Look at key metrics that correspond to your business objectives. For example, if you hope to make more sales, you would examine conversion rate, cart abandonment rate, bounce rate, behavior flows and so on.
The point is to discover what’s working for your site and where there’s room for improvement.
User Testing
Next, you need to discover how customers behave on your site, to find frictions, frustrations, and opportunities for improved customer experience and conversions. You’ll do this through user tests, such as recordings, user surveys, heatmaps, and scroll maps.
With the information, you gather from all of this research you can formulate and prioritize valuable hypotheses for A/B tests.
2. Use Google Optimize
There are a couple of reasons why you might choose Google Optimize over other A/B testing tools.
Firstly, it’s free. So, you can test the value of A/B testing for your e-commerce store before paying for a tool. Furthermore, it’s integrated with Google Analytics meaning it’s an efficient way to create tests using your website data and ‘Goals’
To use the tool, first set up your Google Optimize account, add your container (your e-commerce store URL) and link this account to your Google Analytics property.
To use Google Optimize with Magento, add a snippet of code to your Google Analytics tracking script:
ga('require', 'GTM-XXXXXXX');
Or add the Google Optimize tag using Google Tag Manager.
Then you can start creating experiments within Google Optimize. Click on ‘Let’s go’ or ‘Create experience’ and you’ll see this:
Add the name of your test, the page you wish to test and select the A/B test. Next, select ‘Add variant’ to create a variant page you can edit for your test.
Click ‘Edit’ and make changes directly to your page using the tool:
After you hit ‘Save’, choose ‘Add experiment objective’ on the dashboard.
Create a custom objective or select ‘Choose from list’ and you will see a page like this:
It shows the objectives you may want to test your page for, including goal completions, bounces, page views and session duration.
Now you’re ready to start running your A/B test. As you can see, it’s pretty simple to use Google Optimize on Magento sites.
3. Use an A/B Testing Extension
There are several A/B testing extensions available on the Magento Marketplace.
Popular tools include:
Amasty
Amasty allows you to run several simultaneous A/B and multivariate tests from your Magento admin panel.
All you have to do is choose your goal, e.g. ‘product purchased’ or ‘product added to cart’. Then, test attributes of page design or changes to specific products.
Convert Experiences
Convert Experiences offers conversion optimization tests for product attributes, features and custom designs. The tool integrates with Google Analytics and order tracking.
To use it, tag product attributes and create experiments on the convert.com dashboard.
Mocking Fish
Mocking Fish’s A/B testing tool is free and can be set up on your Magento site using code.
Create experiments by loading your desired page in the Mocking Fish editor. Here you can create variants using the tool or make changes to CSS elements or JavaScript code.
But, how do you choose the best A/B testing extension for your site?
There are several aspects to consider when choosing the right tool, including a budget, skill level, and functionality.
Ask questions, such as:
- How much will the extension cost in the long-term and is the ROI worth it?
- Do you need to know how to code to use this tool and does my team have the right skills?
- What capabilities does this tool have compared to others on the market?
Consider these points carefully and you’ll find the right extension for your e-commerce store.
4. Remove Customer Pain Points
In the competitive e-commerce landscape, a positive user experience makes all the difference. A “superior user experience” could increase conversion rates by up to 400%. Good UX helps customers find the right information quickly and easily and removes frustrations that prevent purchases.
Not only will this lead to more conversions, but also create loyal customers.
User testing in the first stage should have helped you create hypotheses.
For example, “the information architecture is too difficult to navigate which prevents customers from finding the right product.”
Now, you can use A/B testing to test this theory. You might create a variant of your homepage with a simplified navigation menu, to see which menu gets more click-throughs.
Here are some other points of friction to look out for:
Product Search
If customers can’t find items, test a search variant with enhanced Magento filters and product sorting options.
Forms
If customers abandon their items at the checkout stage, test a form with fewer fields.
Too Many Elements
If customers bounce because they’re overwhelmed, try a test where you remove an element, e.g. a floating bar, to see if more users stick around.
Take what you learn from your A/B tests to consistently make improvements to user experience.
5. Test Your Pricing Strategy
Pricing your products is one of the hardest things to get right.
Firstly, it’s difficult to compete with the lower prices of mega e-commerce businesses and marketplaces. Add shipping costs and other fees into the mix and there’s even more difficult. 53% of online shoppers abandon their cart due to unexpected costs at the checkout.
Then there’s the matter of customer perspectives… Price your products too low and customers may think your products are of low quality. Price them too high and customers might think the products are poor value for money. However, there are e-commerce pricing strategies that you could use to boost sales.
All you have to do is A/B test your new pricing strategy against the original price to see if you get more conversions.
Here are some ideas:
Installment Pricing
Offering financing options, particularly the option to pay in installments makes an expensive product seem more affordable.
Casper does it for all of their mattresses:
What’s more, customers may seek out stores that offer installment pricing if they can’t afford to pay for an item right away.
Charm Pricing
This is when you offer slightly less than a round number, e.g. $29.99 or $18.95.
As in this example from Best Buy:
It has a psychological effect on customers, in which they perceive the cost as lower.
Prestige Pricing
If you sell luxury items, rounding the price to the nearest .00 may result in more conversions.
Here’s an example from Louboutin:
The types of customers who buy luxury items want to enjoy the pleasure and prestige of a purchase. They don’t want to bother with complicated numbers. Nor do they want an item that seems cheap.
The trick is to create pricing hypotheses based on your customer base. And then test to see if your hypothesis was correct.
6. Use A/B Testing to Improve SEO
To give you some perspective on the value of SEO…
Organic search sends the most traffic to e-commerce sites (42%), more than paid search (18%) and social (6%) put together.
The cool thing about A/B testing SEO is that you can test improvements without losing any of your existing traffic. For Magento sites, you can test aspects of on-page and technical SEO.
For Magento sites, you can test aspects of on-page and technical SEO.
Here are some SEO A/B tests you can run:
Headlines
Magento automatically creates header and title tags. But to optimize for SEO, you should create titles manually using different product groups.
There are a number of A/B tests you can try for headlines. For instance, placing your keyword at the beginning of the headline to see if it makes a difference to search rankings.
Meta Titles and Descriptions
Amending meta titles and descriptions could lead to more click-throughs from search engines. For example, you could test whether including the word “Buy” in your meta titles equals a better click-through rate.
To alter metadata in Magento, head to your product catalog and click on ‘Edit product’.
Scroll to find the ‘Search Engine Optimization’ section which contains boxes for the meta title and description:
You could also test metadata for category and content pages.
Product Descriptions
Longer product descriptions could lead to better dwell time, more keyword rankings, better page authority and more. For instance, you could test a product description of 100 words against one with 300 words to see what happens to your SEO metrics. Ultimately, you’ll find what length works best for your site.
And small changes like these could lead to higher positions in the SERPs.
7. Test Continually
If you’re looking for continued growth, then you need to continually run A/B tests over time.
David Hoos at BigCommerce says, “Effective A/B testing ideally involves repeatedly testing improvements until you get the best possible version. It’s an iterative process with each test building upon the results of the previous tests.”
As you learn more about what works for your site, take inspiration from guides like this and gain a deeper understanding of customer behavior, you will inevitably formulate new hypotheses.
And as you test these hypotheses over time and make small tweaks to your site, you will see big changes as all of these small elements come together.
Final Word
There’s no doubting that A/B testing results in more conversions. Using it effectively is what’s difficult.
Random hypotheses on what will create more sales for your e-commerce store have no value. You must formulate hypotheses based on research and choose the right tools to test your hypotheses.
The tests themselves should revolve around key areas that stimulate growth, including UX, pricing, and SEO.
And finally, you must continue to run tests and build upon previous tests to ensure sales keep increasing.
Where do you start? Take a look at your analytics and carry out user tests. To come up with the right hypotheses, you need a clear understanding of customer behavior.
About the Author:
Emil Kristensen is the CMO and co-founder of Sleeknote: a company that helps e-commerce brands turn their website browsers into buyers—without hurting the user experience.
Newsletter