The principle behind A/B testing (sometimes called split testing) couldn’t be simpler. You create two or more versions of a page and split your website traffic between each one to see which version performs best.
With each experiment, you test for a single variable so you can accurately measure what impact each successive change has. That variable could be anything from the headline and the amount of copy on the page to the use of images, video or the overall style and navigation.
The granular nature of A/B testing means you can continually refine and adjust various elements to make incremental improvements.
Why is A/B testing important to hotels?
According to ehospitalitytimes, average hotel conversion rates are around 2-3%, so it’s clear there’s room for improvement here. A/B testing can help work out where those improvements can be made.
The continued success of OTAs at gaining bookings can largely be attributed to their investment in user experience, something continual A/B testing has helped them with. In the case of Booking.com, they don’t make a single change to their website without validating it first. Companies such as Airbnb also spend significant amounts of time running tests to improve user experience, including the recent overhaul of their search results page.
While the scale and frequency of these A/B tests might seem intimidating, smaller hotels can make huge strides of their own by testing the areas that stand to deliver them the greatest overall benefit.
Four elements hotels should be A/B testing
The options are endless when it comes to A/B testing. So where to start?
To help define the areas of a website that require the most attention, optimization experts at conversionxl recommend looking at your Google Analytics reports, particularly the User Flow analysis. Getting an overview of where you might be losing customers is a great place to work out where to conduct tests.
Away from existing pages, you might want to test the effectiveness of a new landing page, newsletter or email campaign. When you’re creating new content, A/B testing lets you iron out the creases early on.
Ultimately, the tests you run should be defined by the areas that stand to offer the biggest personal benefits. That said, there are a number of key areas all hotels might want to consider as a starting point.
1. Booking pages
A study conducted by SalesCycle found that 29% of guests abandon online bookings because of how the booking process is set up and executed.
Breaking that figure down, the reasons guests abandoned were as follows:
- 1.53% – When shown the price
- 2.26% – When asked for personal details
- 3.21% – When asked for payment details
How could A/B testing be used to address these issues?
If a significant number of people leave when they see your price, the price itself may of course be the issue. But some customers might be leaving to compare the price elsewhere, or simply need a little more persuasion to book.
If you have a best rate guarantee policy, perhaps it isn’t prominent enough on your site. A simple A/B test might reveal that mentioning your best rate guarantee more frequently, making it larger, or placing it in a more prominent position increases conversion rates.
If people seem to be leaving your site when asked for personal details, testing variations on shorter or simpler form fields could help.
And if asking for payment details seems to be a stumbling block, it might be that customers need reassuring about the security of your site. Carrying out tests that make your trust icons more prominent could have an impact here.
2. Calls to action
Carrying out tests on your call to action can also have a major impact on conversion rates. In the case of French hotel and vacation-booking site Pierre et Vacances, they boosted bookings by 13.6% by comparing and subsequently changing their CTA copy from “Book” to “Check Availability.”
While the specific text of a CTA can alter response levels, its position on a page and visual design can also be crucial.
Changing the CTA button size, color and style of font can all be measured for effectiveness. Focusing on this kind of detail might seem inconsequential, but small changes can lead to huge improvements.
The Vineyard, a luxury hotel near London, repositioned their CTA to the top of their page and swapped a text link for a button, which increased their click-through rate by 32%.
3. Landing pages from SEM/PPC campaigns
If you’ve designed a landing page to support a paid advertising campaign, there are a number of considerations worth looking at to ensure high conversion rates.
Firstly, landing pages should be clean, simple and have minimal distractions. To ensure optimal user experience, consider testing variations on the amount of copy featured on the page, or test alterations on opt-in forms to ensure they’re as easy to fill in as possible.
Bear in mind that if you do alter the copy on your landing page, make sure this change is reflected in your advert. A consistency of message is important for your visitors and also heavily impacts your quality score from Google.
The benefits of your offer should also be clear and easy to understand. Running tests that compare variations on your landing page’s value proposition could significantly alter conversion rates.
To boost engagement levels further, you might also want to include testimonials, images and even video content on your landing page. Again, A/B testing will help you gauge the impact these changes have.
4. Email marketing campaigns
Many of the tests referred to so far can also be used in your email marketing campaigns. But there’s one element of your emails that’s arguably worth testing more than any other: your email subject line.
Ultimately, the subject line dictates whether somebody opens your email or completely ignores it, so it’s well-worth spending the time to get this right.
Try multiple versions of subject lines among your email subscriber list to see which achieves the highest open rates. This needn’t be viewed as just a one-off experiment. The results you get back will help short-term but could also help you spot patterns for future campaigns.
Depending on the nature of the email, you could also different images to test for overall effectiveness. For instance, a package deal combining a two-night stay with a meal at your restaurant could include a general photograph of your restaurant in one email and the chef’s signature dish in another. This kind of test can reveal whether selling the ambience or the food has a greater impact on response levels.
Customer reviews can also be highly persuasive when used in email campaigns. Again, A/B testing could help you find the perfect balance in terms of which reviews you use, the amount you include and where they sit within the email.
Finally, the time an email gets sent out can play a major role in overall response levels. Consider splitting your email list in two and scheduling the same email to go out at different times to see which receives the highest open rates.
Email marketing provider MailChimp offers some useful guidance on this very subject to help businesses work out the optimal time to send emails out.
Test, iterate, then re-test
A/B testing is very much akin to a scientific experiment. You carefully change the conditions each time by altering one variable to see which changes lead to improvements.
You may not have time to test every element of your website or marketing campaign, so hone in on the areas that are going to make the biggest difference. Essentially, measure what matters.
Above all, see A/B testing as an ongoing process. The OTAs have managed to gain more bookings because they see view continual testing as an essential way of constantly iterating and improving user experience.
In the same way, hotels need to see A/B testing as just another part of their business. Not only can this help bring about more more direct bookings, the very process of continual testing reinforces a general mentality of constant adaptation to shifting customer needs.