Case Study: How our integrated experimentation team fueled a homepage refresh

March 03, 2025

In March 2022, Valtech wrapped up a digital transformation project for a company in the food and nutrition space, culminating with the launch of a fully redesigned corporate website.

As expected, the homepage received a lot of attention during the redesign process. There were multiple discussions about the best layout for the homepage and how best to showcase the breadth of services and solutions the company offers. This sparked the all- to-common question that comes up during these debates: how we showcase everything a company does succinctly without creating an overwhelming user experience.

The compromise ended up being a homepage carousel/slider. While this was not our recommended approach, it allowed us to get the buy-in we needed to move the project forward.

However, our early post-launch analysis revealed that users did not interact favorably with the carousel.

Designing the A/B Test: Homepage Hero Carousel Pilot

Six months after launch, we conducted in-depth analysis of the overall performance of the website. Some of the early trends were striking: the company's homepage accounted for nearly half of all site entrances, but over 50% of visitors bounced without engaging with the page. Even more telling was that less than 1% of visitors who landed on the homepage engaged with the CTA buttons in the carousel.

We turned to our experimentation program to try to find a solution.

To address this, we initiated a pilot A/B test using Optimizely Web Experimentation.

At the end of the A/B test, we had the insights we needed to make informed decisions on how to move forward.

Condensing the hero banner led to a 13.5% increase in CTA clicks and a statistically significant rise in total hero banner interactions. Despite this, fewer than 1% of homepage visits resulted in CTA button clicks.

Moving Forward: Homepage Redesign Process

While the Variant had a significant impact that generally supported our hypothesis, it still resulted in less than 1% of CTA button clicks, indicating there was clear room for improvement. As a result, we recommended a full homepage redesign.

The A/B test results gave our internal stakeholders the ammo they needed to sell this homepage redesign internally. With full buy-in across the board, our integrated experimentation team got to work.

This process began with the research phase, which included:

• UX Research: We provided UX best practices and conducted studies on homepage carousel interactions.

• Competitor Research: Analyzed four key competitors' homepages and proposed best practices.

• Persona Journeys: Created three user personas to map pathways needed on the homepage.

• Behavioral Analytics: Analyzed user engagement with homepage components, revealing low interaction with the carousel and other elements.

• Search Trends: Found that visitors often searched for the company's locations and jobs, indicating a need for clearer information about the company on the homepage.

We presented three new homepage hero components and a personalization component that can be used for specific campaigns. The proposed dynamic header aimed to provide enough context for general users while directing targeted users to relevant content.

We provided data-driven decisions on each component to include on the proposed homepage. The components were strategically ordered to tell a story about the company and everything they do.

Preliminary data and user testing results guided the first homepage restructure design. Some of the key insights from user testing on Maze included:

Positive Outcomes:

• Increased user understanding of the main message.

• Clear headlines and CTAs were well-received.

• 19% more users could recall the heading.

• Updated components without carousels or hover effects improved section comprehension.

• Positive emotional response to the redesigned homepage.

Areas for Improvement:

• Users struggled to grasp the hero content at first glance.

• Messaging needed to be more concise and clearer.

• Some images affected users' understanding of the company's purpose.

• Needed more intentional selection of images and headings.

After iterative design adjustments based on user testing and data, the new homepage was launched. Subsequent user trunk testing showed overall improvements and positive data trends within the first month.

The Results

After one month of the new homepage being live, the improvements were clear: engagement metrics improved overall, and more users were interacting with the content of the page.

The journey of a food and nutrition company's homepage redesign highlights the importance of user-centric design and iterative testing. The initial analysis and A/B testing paved the way for a comprehensive homepage redesign that significantly improved user engagement and understanding, demonstrating the value of data-driven decision-making and an integrated experimentation team.