A/C Testing for Headless CMS – Traditional CMS
is a reliable paradigm, but recently headless content management system or (CMS) has gained popularity because of its flexibility.
If you’re running an e-commerce company, it is helpful to reuse the same website content for other digital platforms.
Traditional CMS focuses purely on publishing on the web, while headless CMS setup is more flexible in creating, publishing, customizing, and integrating your content.
So, if you want to increase conversion rates, this guide will show you how conducting
AB testing for ecommerce can help you optimize your website pagin no time.
I run an eCommerce site, have headless CMS, and want to setup AB testing (or split testing)… now what?
- Did you identify and decide on your goals?
- Do you have a way to measure and quantify them?
- Do you have a way to monitor said metrics?
- Do you have a way to automate the work for similar future changes?
- Can you delegate some or all of the work on split testing to another business?
- At what cost? What will be the time spent?
- Can the decision be reversed?
- How future-proof do you need the solution to be?
How To Do A/B Testing For Headless CMS?
Since it’s decoupled from the frontend, headless connects the backend to other digital platforms through an API.
As a result, you can publish your
content so customers can see the page on any device.
- Your web page elements will be easy to see and navigate from people’s mobiles or desktops.
- Your apps will be updated instantly with the new content.
- Your subscribed visitors can even get notifications to sign on using their smartwatches!
If you have an e-commerce business, you will minimize the use of resources such as CPU, database operations, caches going cold, etc.
At the same time, try to gain more conversions by offering bundles, showcasing related items, and allowing people to compare items. Conversion rate will increase when you find the perfect mix of elements that customers prefer to see on a web page.
So, AWS (Amazon Web Services) publishes an SDK (Software Development Kit) for every service. This SDK is correlated with every declared API, so it enables ease of use.
So what makes A/B testing for headless different? Before you jump into it, be sure to check out our guide on
how to prepare for AB testing.
The metrics for headless A/B testing are more focused on what your CMS controls vs. on-page changes.
If you measure a new feature for a product or the best form of intake using your headless, these are some sample data metrics you should track:
- Revenue
- Transactions
- Error messages reported
- Bounced Sessions
- Average order value
That leads us to creating site variations through the use of A B tests:
- Version A: an input field for product (free form text)
- Version B: a drop-down with the available products when you select a category (like “dresses” or “shoes”)
How many customers will use the product field to identify the product they’re searching for? Is there any difference between each variation?
Golden rule: Ask yourself, “what is the goal of this feature or change?”
- Increase number of purchases
- Increase AOV
- Increase add to carts
- Decreases the user bounce rate
- Reduces the number of returns
Basically, you want to increase profit while decreasing cost – that’s the bottom line.
The twist is it’s not the same interface between the front end and the backend of your site. The backend publishes contracts independently from the interface – these data should ideally be backward compatible so they don’t crash and burn on your web pages.
So basically, there are two options:
- The backend of the site publishes two separate variations. E.g., If I add a field with a particular type of behavior does it increase the likelihood of the user converting?
- Change one contract in the input to observe how variables change.
A headless content management system is also relatively straightforward to use with A/B tests and employing different software for testing.
For example, if you are testing search tools like Algolia vs. Twiggle, having headless can make setting up the test easier than getting Magento to serve up different search integrations alternatively.
The test would still require development work, to be sure, but a headless, API-driven approach might make it easier.
Why Is It Important To Run Experiments On Marketing Content?
Marketing tests look at various features and functions to ensure your site visitors and customers have the best possible experience. Of course, that experience has to end with them purchasing your products, so your website has to have a precise sales funnel.
Integrate A/B Testing With Headless CMS
Some tools like Google Optimize allow you to seamlessly implement A/B testing for headless if you know some basic HTML. You’ll set up the test in minutes, and your customers won’t have a poor experience because of that test.
Next, you’ll have to wait a few weeks before getting the results and implementing the changes.
Pro tip: A web page must have enough data- at least 10,000 visitors worth- for the test to be relevant.
And there’s another thing:
Not everyone will know what to test. Sometimes, it’s as easy as moving the CTA up or down on a product page. Other times, it could be changing the background color of a particular product, adding images, changing the layout, or minimizing text to decrease bounce rates. It’s all about finding the landing page that customers prefer to see. You will want to do multiple test runs over the course of a few weeks or even months in order to find it.
That means, in most cases, A/B testing with headless CMS is best left to the experts.
Please check out our guide on why you should
outsource AB testing for ecommerce. Some SEO agencies integrate this sort of service into their offer so that you can get a discount for the whole package.
Tools like
this are very relevant. However, if you don’t have someone on your team who can do it, you’d have to estimate 2-3 months to hire and train someone.
Alternatively, if you want to get into it yourself, below are some:
A/B Testing Tools And Their Reviews
A/B Testing Tools for Traditional CMS
1. Google Content Experiments: Free
Image source
One of the advantages of Google Content Experiments (Formerly known as Google Optimize) is that it is free. Another powerful feature is its integration with Google Analytics.
For example, you can use Google Analytics data to spot critical segments of site users to include in your target audience.
Once you’ve identified your key audiences, you can create a unique offer for each target customer or user group and then use data from Google Optimize to tailor the offer.
2. Optimizely: Best for Enterprise
Image Source
Optimizely is one of the most expensive tools out there. So think twice before deciding to go for it.
The reason is that Optimizely targets enterprises with significant website traffic. It says that most tests are run at 90% statistical significance. As a result, it might not always work because even the websites with massive traffic sometimes struggle to reach well-oriented statistical significance.
If you have a site with low traffic, you may want to run a risky test or two that need 90% significance to have a valid result.
However, it is difficult to test accurately with low traffic. So consider free or less expensive tools to conduct your experiment.
3. VWO: Best for Small Businesses
Image source
One of the advantages of VWO is flexible pricing. It allows you to pay by the average number of unique monthly visitors. As a result, it doesn’t matter how many domains you run tests on as long as you are within the visitor quota.
VWO doesn’t have a central “AB test goal management” section inside their AB testing product, making it different from Optimizely.
Instead, you can see your most frequently used goals as a list when creating them for any test you’re running. Cost is the most significant factor that makes VWO ideal for small businesses.
A/B Test Tools for Headless CMS
1. Kentico Xperience: Best for Startups and the Non-Technical User
Image source
Kentico Xperience is best-rated for ease of use, features, and customer service. We recommend it for startups because it’s scalable, allowing you to test further as your website grows.
Plus, Kentico works well to develop niche new content type for your target audience.
Use this tool to edit your product descriptions and CTAs to improve your sale rate or prevent checkout abandonment.
2. Contentful A/B Test: Best for Short Text Edits
Image source
Contentful A/B testing platform is one of the most popular choices for marketers because you can use it to edit short texts or paragraphs quickly.
Short texts are precisely what
you have on your website – whether on your product or checkout pages.
Also, this platform boasts a wide range of languages for content creation and feature-rich APIs for system integration. As a result, you can unify all your content in one hub.
So, when you add something new to your webpage, you can immediately test how that same feature works for mobile phones.
3. Contentstack A/B Test: Best for Ongoing Support
Contentstack is best if you’re working for a budget-conscious company and want prompt support from a knowledgeable team.
For example, if you’re not exactly sure how A/B testing works for headless, Contentstack is an excellent place to start because the entire team will be there by your side.
Pro tip: You should also choose Contentstack if you’ve recently transitioned to a headless eCommerce website.
Apart from the expert support, Contentstack allows you to upload all the product data to their platform. That means fewer API calls on your web page so that your website will always run at a high performance.
So, if you’re busy with this transition, Contentstack will improve your entire team’s workflow.
4. Sitecore A/B Test: Best for Accurate Tests
Image Source
Sitecore’s unique selling proposition is integrating A/B testing tools with Path Analyzer, a tool that thoroughly assesses and provides insight into your website’s performance.
That means your A/B test will focus on the right target from the get-go.
For example, you can comparatively test multiple variables that mess with your checkout process. You could check whether your customers want a more straightforward checkout, more trust seals, or alternative payment options.
Sitecore’s algorithm will evaluate if a test is or isn’t working well. That way, you’ll get accurate test results and know what changes to implement to increase the conversion rate.
Final Thoughts
An A/B test evaluates whether the current web page requires changes to attract more users.
That’s why A/B testing is essential if your e-commerce store is using headless CMS.
Remember that you’re going to have the same content published on many platforms simultaneously. So even though your website’s perfectly intuitive, user experiences on the app can falter because that content is on two different channels.
You can choose one of the five tools we discussed above to make sure everything runs smoothly for your customers.
An A/B test for headless requires that you test different variations against the original version, so you need the expertise to get accurate test results.
That’s why, for novices at least, we recommend reaching to a specialized agency first.
