Skip to main content
a/b tests

A/B tests for different kinds of businesses (e.g. eCom, SaaS)

December 7, 2021
Reading time: 
15 mins
Daniel Boltinsky
Daniel Boltinsky
Kameleoon, Managing Editor, North America

Like using the scientific method test our hypotheses about the natural world, A/B tests remove the guesswork about what your audience responds to best. A/B tests provide definitive, empirical results so you can make changes to your website and product that drive more revenue.

Whereas changes based on gut feeling can hurt your business, A/B tests allow you to roll out changes progressively to validate your site’s pain points and use data to eliminate them.

But while a culture of experimentation will help any business grow, certain kinds of A/B naturally lend themselves to different types of companies.

For example, eCom can multiply their revenue by running A/B tests to reduce shopping cart abandonment, whereas software companies can perform pricing experiments on their pricing plans to see which features customers do and don’t want.

A/B tests for online stores

1. Experiment with copy to validate what your customers want

When customers land on your web page or any other piece of your marketing material, they should feel like you know them and understand their needs.

Jerry Han, chief marketing officer at PrizeRebel, agrees and says, “Effective messaging means that you can demonstrate the value of your product and convey exactly why the user should take a certain action.”

Effective copywriting is a well-researched hypothesis about what you think will convert lookers into buyers. And the point of testing it is to see if your hypothesis is correct.

Han goes on to say that if you find that your conversion rates or impressions are lower than you’d expect, your website copy could be the culprit. A/B testing your website copy can help you identify whether you can improve messaging elements.

Effective copy experimentation does not mean changing a few words to try to influence a user subconsciously.

Experimentation and copywriting work together effectively because you have no way to know the real benefits of your product or service unless you test them. By experimenting with different narratives, different testimonials that focus on various product benefits, you can double or even triple your clickthrough rates.

2. Increase conversions by testing page headlines

Much like when you enter a room and perform a scan to do a quick inventory, readers scan a page to decide if the information is worth sticking around for.

And while compelling copywriting is important, headlines and subheadlines—commonly referred to as H1, H2, and H3 tags—are where their eyes will rest first and signal whether or not they should stay on a page and take further action.

Michael Alexis, CEO at TeamBuilding, discovered the following after A/B testing headlines: “The most impactful A/B tests we’ve run are on H1 and H2 page headlines. The reason these headlines are so important is that the more clear they are, the more likely people are going to continue reading and then take some measurable action. By removing the guesswork, you can almost always increase your conversation rates.”

A/B testing is a way to hypothesize, test, and validate what it is that your customers actually want. Knowing what your audience needs from your product and using that knowledge in the headlines will increase the likelihood of a conversion.

3. Personalize your calls to action (CTAs)

CTAs are the gateway to conversions. But it’s more than just the color or size of the button.

Some marketers, like Josh Brown, the head of content at Helpjuice, like to optimize CTAs based on where the user comes from and personalize the CTA accordingly. He suggests considering things like traffic source, user intent, page content, etc., in order to create the right type of CTA that has some personalization aspect to it.

Brown says he also likes to optimize CTAs with A/B tests because at the end of the day, that’s pretty much the entire point of getting traffic to your website—it’s to get users to take some action, whether it’s to sign up for your newsletter so that you can build a relationship (and then convert to a sale) or immediately get the user to make a purchase.

You can test everything from size and color to text and button position to determine what instigates the most clicks.

4. Use experimentation to decrease abandoned carts

Your customer’s shopping cart experience starts the moment they add an item to their cart and ends when they receive the final product (kind of, but that’s another article), yet some customers never even make it to the checkout page. With cart abandonment rates at a whopping 84.27% in ecommerce, the checkout experience is one of the most critical elements of your website. If a customer makes it all the way to the checkout and then jumps ship, it’s imperative to find out why.

For reasons that you’ll never know unless you test, customers drop off at various stages during the journey of add to cart to checkout, and understanding what elements can be changed to decrease the number of abandoned carts can help boost sales.

Mike Nemeroff, CEO and co-founder of Rush Order Tees likes to test the number of steps in the process and has found that fewer isn’t always better, especially when you consider adding special offers, cross-sales, and other value propositions.

He suggests looking at the confidence builders you’re using, like social proof, customer quotes, security labels, and CTA buttons. He notes all of this can have a marked effect on what you’re doing at the most vital sales process step. Any of these functions can and should be tested to see what gets more customers to enter their payment info and complete the checkout.

Optimizing your shopping cart can result in more customers completing orders (higher conversion rates), and the best way to do that is to review your shopping cart analytics.

Joe Brown, founder and digital strategist at Digital Junkies, agrees, citing, “Cart abandonment indicates that there are issues with the process after adding items to the cart. Once you fix the issue, all those abandoned carts will turn into your sales, removing a massive hurdle from your business’ success.”

Many factors can contribute to abandoned carts, like unexpected shipping costs, lack of payment options, and unclear shipping dates and carriers. If you don’t review the stats, run A/B tests, and compare the results, you'll never know what they are.

5. Discover which image style has the greatest impact

If the adage “a picture is worth a thousand words” is true, then the primary photo on your product landing page is worth a lot of money.

Depending on the type of product you sell (electronics vs. clothing, skincare vs. luggage), the first photo that loads automatically on your product landing page (PLP) will help the customer determine if this is the right product for them.

Founder of digital marketing agency GR0, Jonathan Zacharias, believes one of the most important and vital A/B tests for ecommerce stores includes testing a primary lifestyle versus product shot on a website’s PLP.

This primary photo can be a lifestyle photo (the product being used by a real person, like someone wearing a bathing suit on the beach) or a product photo (the product alone on a white background, for example).

Zacharias urges ecommerce owners to test these two variables by making both standard lifestyle images and standard product images. He says each photo should have the same angle, perspective, and editing, therefore stressing uniformity and professionalism to the customer.

Test both for three to four weeks and measure the click-through rate (CTR), conversion rate, and add-to-cart rate accordingly.

A/B tests for businesses running digital ads

6. A/B test variations of your copy for greater ROI

In digital ads, the copy is often the second most influential part of the ad itself (the image or video being the first), and so it’s important to test the copy to see what resonates most with your target audience.

“Without good marketing [ad] copy, you won’t be able to attract the attention of your target audience,” observes Kristen Dole, director of growth hacking for Nowadays, a creative agency.

“Luckily it’s easy to perform A/B testing for marketing copy through paid ad campaigns on platforms such as Facebook. A/B testing for ad text will be very helpful in order to determine the kind of marketing copy that helps you acquire more leads.”

Testing your ad copy means only testing your ad copy (and not the other elements of your ad). This ensures you get very clear results from your tests on what messaging your target audience responds to. You can do this by writing different versions of the same ad — this can mean different headlines but the body remains the same, or different body copy but the headlines change. These types of tests will inform on what headline and what body copy perform best and get you the most clickthroughs from the ads and the most purchases. The other type of copy you can test for your ads is the CTA. The words on that button can make a big difference — and you’ll only know that if you test it.

Just like the copy on your website, you’re just hypothesizing what copy and messaging will lead to conversions. This means isolating and testing your ad copy is the only way to know if it’s driving conversions. More clicks on the ads will translate to greater conversions, and that means greater ROI for your digital ad spend.

7. Select the right keywords for pay-per-click (PPC) ad campaigns

Since most PPC ads are powered by your keyword selection, it makes sense to optimize for the best keywords for your target audience, even when running several campaigns simultaneously.

At Home Buyer Louisiana, Stephen Keighery, CEO and founder, says he runs their campaigns for long periods at a time, so it’s crucial they get the most out of their budget.

They perform A/B testing by designing multiple campaigns—each with unique keywords—before running them simultaneously over the course of six months. The results of which inform their next campaign and other marketing initiatives.

Absolute Reg LTD is among thousands of businesses that use paid Facebook ads to promote their business and A/B testing to help find the best campaigns in terms of ROI.

Jake Smith, managing director, reports, “After the [test period] is over and the results are in, we look at the ROI of each campaign so that we can determine what works best for our audience. … From there, it’s a matter of replicating the most successful campaigns and updating them when necessary.”

So how do you go about selecting the right keywords for your digital campaigns? When it comes to selecting the right keywords for PPC ads, Google says it best: “The right keywords can get your ad in front of the right customers.” Their nifty Google Ads Keyword Planner helps you not only pick the best keywords to target for your campaign, but also provides you with bid estimates for those specific keywords. Do you prefer to spend X on a high-volume search term or do you prefer to spend Y on a lower-volume but more specific keyword?

This point brings us to how you can incorporate A/B testing with your keyword selection. The first option is to test the keywords themselves to see which terms bring you the most qualified leads. The second option is to test different versions of the same ads (as we mentioned before) but for the same keywords.

A/B tests for software companies

8. Use price experimentation to determine the best software plans

Software as a service (SaaS) companies often have different pricing tiers for their products as well as either a free trial of the product (sometimes for 14 or 30 days) or a free version of the product (often referred to as a freemium). Testing these pricing schemes can show what consumers are willing to pay, as well as determine whether a freemium or free trial is a more effective option.

This sentiment is confirmed by Scott Keever, founder of Scott Keever SEO, who says, “using A/B testing to find the best pricing scheme [is important] because that way you can choose between time-sensitive freemiums, free trials, or money-back guarantees to assign the best product demos.”

Keever adds that A/B testing can also help marketers decide the length and time of every free trial and, later, the final set price for the product or service. These tests help consumers sort through different bundles and assess the price tag concerning the value consumers feel from the offering. They help to know the perfect number of days for the product demos and the pricing bundles.

9. A/B test the ideal number of form fields

Whether it’s for a free download, gated content, or to sign up for a new bank account, the number of form fields you require could have a huge impact on your conversion rate.

According to Oli Gardner, co-founder of Unbounce, having from one to two form fields results in 15.11–17% conversion and drops significantly when you have from four to seven form fields (down to 7.18–7.58%). With such a small difference between four and seven fields, Gardner recommends testing the impact of four fields vs. seven fields.

Matt Weber, founder of Weber & Co., knows firsthand the impact of A/B testing the number of form fields. After testing this, he found the form with fewer fields had a 20% increase in conversions. He expressed a key takeaway from this A/B test was that marketers must pay close attention to all aspects of the marketing funnel because small changes can have big impacts on an organization’s bottom line.

A/B Tests for Financial Services

10. Optimize landing pages for financial products

For financial services, landing pages are arguably one of the most important elements in lead generation. Getting these right can mean the difference between a conversion and a bounce. If your value proposition isn’t explicitly laid out (i.e. they don’t convince your site visitors to take action), you will lose the lead.

This is where A/B tests come in.

“On the first landing page in our account creation funnel, we had a lot of traffic but hardly any conversions,” explains Justine Stevens, digital marketing manager, Hello bank!. As an online direct bank owned by BNP Paribas Group, Hello bank! needed to get this right. By creating different hypotheses, Stevens and her team created several variations (six in fact) since this page was particularly important to new customers to sign up. The A/B tests they conducted saw a 23% growth in accounts being created.

Banks and other financial institutions rely on customers creating new accounts, so being able to identify troubled pages and create working theories on what the problems are based on analytics and testing can help you increase your conversions.

A/B tests for Healthcare

11. Testing banner ads on your own landing pages

When you think of banner ads, you think of digital advertisement on third-party sites. But that isn’t always the case. You can create banner ads for your own site to promote your own initiatives.

In the case of Humara, a healthcare insurance provider in the U.S., they were having low uptake in one of their programs that provides those in the healthcare industry with social media insights.

Through testing, they determined that their audience responded best to the landing page with no banner. Again, they could not have determined this without developing a hypothesis and testing it against the control (the original).

This goes along with the findings from the Forrester Opportunity Snapshot that concludes that healthcare organizations (HCOs) that have a digital-first marketing approach are surpassing their competitors with better customer experience (CX).

Leaders responsible for marketing and CX at HCOs recognize that data, analytics, and the insights that flow from them play a pivotal role in ensuring that healthcare consumers have positive experiences.

Seventy-two percent of healthcare organizations surveyed use A/B testing as a personalization method to constantly improve and optimize experiences for their customers.

The Forrester report also suggests, “Managers should turn to profiles and recommendations to make the most of each customer visit and testing to answer why a particular interaction is valuable.”

HCOs can drive personalization with the use of segmentation and data and by leveraging personalization software, you can have advanced testing capabilities. For those worried about remaining Health Insurance Portability and Accountability Act (HIPAA) compliant, you can safeguard patient data while still running A/B tests and other forms of personalization.

Get started with A/B testing

A/B testing is an important function of running an online business since it can help you optimize your work while increasing conversions. To see how easy it can be to run A/B tests on your website, book a Kameleoon demo.

Topics covered by this article
Daniel Boltinsky
Daniel Boltinsky
Kameleoon, Managing Editor, North America
Bold Predictions Live Session
Recommended articles for you