Case Study: Multivariate Testing for CTA Buttons

February 7, 2019



A travel booking agency for young professionals.


Website data showed the customers were getting stuck on mobile devices at the tour info page and not moving forward to get quotes. The goal of this project was to get more customers to get a quote and book a trip.


The brand used SiteSpect to A/B test three different factors on the tour info page: color, position, and text. The colors tested were black versus blue buttons. The positions tested were lower left versus centered. Different variations of the text included: “See dates and prices”; “Dates and prices”; “Get a free quote”; and “Get dates and prices.” These factors resulted in 16 different variation groups. Underperforming combinations were turned off periodically throughout the length of the campaign to allow more traffic to those with positive changes.


The winning combination resulted in a 6% increase in clicks to the “See dates and prices” button and a 5% increase in clicks to select a date and receive a quote, both at 99.9% statistical significance. Total bookings had a 4% directional increase, equal to 300 more bookings during the A/B test for this variation group. This brand is now running a follow up A/B test that is building on the winning changes of the first A/B test and making additional changes to color, font size, and CTA text.

To learn more about SiteSpect, visit our website.


Kate Orchard is a Manager of Customer Success at SiteSpect, where she consults SiteSpect users on their optimization and personalization road maps and projects. She is based in Boston.

Suggested Posts

Hand putting wooden five star shape on table. The best excellent business services rating customer experience concept

5 Ways to Use Product Recommendations for a Better Customer Experience


Using CNAMEs to bypass ITP has been put to Torch

SiteSpect dashboard

Blogs: Quill Enhances its Check-out Experience with SiteSpect

Subscribe to our blog: