Two Lane Design Process: Experimentation & Comprehensive Strategic Design Discovery
Product Design • Experimentation • Two-lane design process
Challenge
HelloFresh customers struggled to find and select meals efficiently due to a lack of structure in the recipe catalog and missing navigation tools. The experience required too much time and effort, impacting user satisfaction.
Approach
To address both short-term usability and long-term structural improvements, we ran a dual-track design process: quick UI enhancements (filters and sorting) alongside in-depth research into meal selection drivers and information architecture. Usability testing and A/B experiments guided iterative refinements.
Outcome
Implemented UI improvements led to increased engagement, with filter adoption rates up to 40% and net revenue gains in key markets (+0.4% to +1.4%). However, A/B tests revealed that additional tools didn’t always increase meal selection. The project initiated ongoing collaboration with operations teams to refine the taxonomy and align internal tools with user needs.
My Role
Sr. Product Designer
Methods
Remote Testing, prototyping, A/B testing,
Year
2022 - Hello Fresh
Design Rationale
How We Started
HelloFresh customers frequently struggled to find and select meals from their weekly menu. User feedback highlighted that meal selection required excessive time and effort due to an unstructured recipe catalog, a long list of options to scan, and the absence of effective navigation tools such as filtering and sorting.
Our key problem statement: The recipe catalog was not designed with the end user in mind.
To address this challenge, I led a structured design process that balanced quick UI improvements with long-term strategic initiatives.
Dual-Lane Design Process
To ensure a comprehensive solution, I mapped out an opportunity framework and convinced stakeholders to adopt a two-lane design process:
Discovery Lane: A long-term initiative to define the best information architecture (IA) and enhance recipe taxonomy.
Experimentation Lane: Quick UI enhancements, such as filters and sorting, to improve immediate usability while gathering user insights.
Understanding Meal Selection Drivers
To build an effective filtering and sorting system, we first needed to understand what truly influences customers' meal choices. A deep dive into academic research on meal drivers revealed a critical gap: Taste was missing from our selection criteria, despite being a key factor in decision-making alongside Nutritional Value, Convenience, and Popularity.
UX Research & Testing
Conducted remote usability tests with 16 U.S. customers to assess the effectiveness of new filtering and sorting tools.
Introduced A/B testing post-implementation across multiple markets (US, AU, DE) to measure impact on net revenue.
Engagement data showed customers preferred collections over sorting, with CTR rates of 29-40% vs. 4-7%.
Outcome & Business Impact
The A/B testing results showed a net revenue increase of +0.4% (US), +1.4% (AU), and +1.1% (DE) for the winning variant.
However, data suggested that introducing new selection tools did not necessarily increase meal choice rate and, in some cases, even decreased it.
The research validated that meal selection is influenced by deeper structural issues in our recipe taxonomy, requiring a long-term solution beyond UI fixes.
Next Steps
Continue working with Operations to refine our recipe taxonomy and ensure it is operationally feasible.
Collaborate with internal product teams to align tools used by menu creators with customer expectations.
Keep iterating and testing, ensuring every improvement is data-driven and enhances the customer experience.
