How I boosted sales for a struggling product
UX Designer
March 2021 - May 2021
Conversion Rate Optimization Manager,
VP of E-Commerce,
Senior UX Designer,
Copywriter,
Graphic Designers,
VP of Brand,
Website Project Manager,
Art Director,
Business Intelligence Manager, and
E-mail Marketing Specialist
Business Goal: Sell the Grill and keep inventory moving to make profits and expand the product line.
User Goal: Enjoy quality time with self, friends, or family by a fire.
The company struggled to sell its charcoal grill, and idle inventory costed money and warehouse space.
Leadership decided to launch a Memorial Day campaign highlighting the Grill. As the UX designer, I was tasked with optimizing the conversion funnel by improving the promotional landing page's clickthrough rate so that more customers would complete their purchase on the store website.
Between major campaigns, the Grill landing page performed drastically poorer compared to the landing pages of other products.
I researched the top charcoal grills of 2021 and discovered that Solo Stove's Grill was marked at a much higher price and offered minimal features compared to other grills. It retailed at $774.99 when popular grills at the time sold for less than half.
While the landing page was not technically a product page, I studied the Baymard Institute's articles to learn how customers understand and interact with product content.
The biggest takeaway was that customers need factual, basic information about the product before they can appreciate its features and highlights.
The CRO Manager and I looked at user analytics in Heap and found that 20% of past purchasers had visited the Grill's product details page in addition to the landing page.
It was as if customers relied on the product page to validate what they gleaned from the landing page.
Considering the research thus far, I hypothesized that users will engage more with a landing page that resembles Solo Stove's product details page, not an ad. The landing page would need to be a trustworthy destination that prioritizes and highlights essential information over marketing claims.
I reviewed the existing landing page against Baymard's best practices, accessibility, and usability heuristics. The Brand team had designed this page, so when ownership transferred over to me, I identified several problems:
I also wanted to get potential customers' points of view and recruited participants on UserZoom. I asked them questions focusing on product education, particularly on whether or not they could identify reasons for buying the Grill.
over the long chunks of text.
explain what made the Grill better than others.
about the lack of concrete details.
The research findings validated the assertion that customers need concrete details before they can appreciate product highlights.
The Baymard Institute's studies offered much guidance on how to structure information and support customers' understanding:
I introduced mobile-first to Solo Stove, since 70% of website visits came from mobile devices. These wireframes mimicked the structure of Solo Stove's product details page, which is why I did not draft multiple iterations. In addition to design, I was also responsible for building and launching through Unbounce--a landing page tool--so I needed to create a design that I could feasibly implement on my own. Once stakeholders approved of this direction, I directed the Brand team in providing copy and image assets.
It was awkward presenting to the VP of Brand--the stakeholder who directed the original landing page design--but I highlighted 3 main rationales for the design changes.
The VP of Brand admitted that the previous approach lacked content strategy and that I infused clarity into the design. All stakeholders approved it for launch.
The two pages--the old page as the control vs new--were A/B tested over a two-week period using the landing page builder Unbounce. I achieved the goal of bringing more users onto the site, demonstrating the value of research, testing, and user-centered design.
Accomplishments:
Home
© 2025 Jason Wong