Travel Insurance
Comparison feature development
Overview
Project summary
Led the end-to-end development of a comprehensive comparison feature for SingSaver's Travel Insurance vertical. The goal was to empower users to make informed decisions by facilitating side-by-side comparisons of complex insurance products, thereby enhancing user experience and increasing conversion rates
My Role
As the Lead Designer, I spearheaded the project from initial research through to launch. My responsibilities encompassed identifying user pain points, formulating hypotheses, designing solutions, collaborating with cross-functional teams, and overseeing the implementation to ensure alignment with user needs and business goals.
Background and Context
Platform
SingSaver is a comparison website best known for Credit Cards sign up offers and enticing rewards. However, they also have other product offerings such as Insurance and Personal Loans which are much more complex. This feature was designed for the Travel Insurance vertical specifically as it is the second highest revenue vertical, right after Credit Cards.
Travel Insurance as a vertical
Compared to Credit Cards, Travel Insurance is a very needs-based product where users will buy travel insurance when they travel. At the same time, purchasing Travel Insurance is can get confusing and even frustrating for users with various tiers offering different levels of coverage under a single insurance provider.
Gaps and Hypotheses
Gap #1: User drop-off
Issue: Following the results page (RP) revamp, while Views to Apply Click (VTA) improved, we observed a 25% drop in Apply Click (AC) to Lead conversion and a sharper decline in AC to Policy Sold rate—from 40% to 22%.
Hypothesis: Users may be clicking "Apply" prematurely, before they feel ready to commit. The product cards only display four main coverages—those found most relevant in user research—but many users likely have additional, specific concerns not addressed in this limited preview.
Gap #2: Limited comparison experience
Issue: As a comparison site, our goal is to help users confidently choose the product that best fits their needs. However, the current RP design forces users to scroll up and down repeatedly to compare product features, making the experience cumbersome.
Hypothesis: This friction may lead to drop-offs. Introducing a side-by-side comparison feature can streamline the decision-making process and better support users in evaluating multiple products at once, increasing engagement and conversions.
Gap #3: Lack of provider page
Issue: High search traffic for provider-specific keywords suggests users often seek information on a specific insurer. However, users clicking these links are directed to a generic results or lead capture page, often focused on rewards, not the provider’s actual product details—leading to poor user experience and lower conversions.
Hypothesis: Many users already know which provider they want and simply seek confirmation. A dedicated provider page consolidating key details would cater to these high-intent users, reduce comparison friction, and improve conversion rates.
Solution
Optimising the mobile comparison experience
With ~60% of users browsing on mobile, delivering a seamless mobile experience was a priority. We had previously tested a tab-based comparison model, but user research revealed it forced users to "memorize" coverage values as they toggled between plans—leading to frustration and decision fatigue. To solve this, we redesigned the comparison table to enable horizontal scrolling, offering users a true side-by-side comparison on smaller screens.
Accounting for all product attributes
Travel insurance plans often come with complex coverage details, including nuanced sub-limits. While top-level limits may appear similar, critical differences often lie in the fine print. To address this, the new table includes expandable sections and detailed sub-limit views, allowing users to locate specific coverage types and compare them in depth—helping them make confident, informed decisions.
Keeping the important things reachable
Given the sheer number of attributes, especially on mobile, tables can become lengthy and difficult to navigate. To maintain clarity and usability, we designed the plan name, final price, and call-to-action (CTA) to remain sticky as users scroll vertically. We also optimized the scrolling view by reducing font size and padding in that state, ensuring users could view as much content as possible without compromising readability.
Deliver
Defining acceptance criterias
I defined clear acceptance criteria and created detailed designs covering key UI states, edge cases, and user flows. These were handed off to engineers via JIRA tickets, complete with annotations and specifications to ensure accurate implementation. This approach helped align expectations across teams and minimized ambiguity during development.
Roadblocks and solutions
During development, a critical oversight in specifying the search interaction led to a mismatch between user queries and displayed results. Specifically, the comparison table was initially configured to display only single-trip plans, even when users searched for annual plans, resulting in empty or irrelevant results. Recognizing this, I collaborated closely with the engineering team to refine the search parameters and ensure accurate, dynamic content rendering based on user input.
Next steps
UX success metrics tracking
Reduce premature AC / increase AC Quality: Measured by comparing the AC to Lead of Provider page to that of the Results page. Reduced post-AC Dropoff = Improved AC Quality.
Opportunity: Adapting the table to be used for other verticals if the quality of AC is improved.
User interest in coverage information: Track user interaction with the table accordions for sections and coverages.
Opportunity: User segmentation based on coverages that they consider and movement of sections up/down based on user interest to further reduce user effort.
Key takeaway
Design doesn’t end at handoff
Owning the full design process taught me the importance of staying closely involved beyond handoff—especially in collaborating with engineers and QA to ensure the feature worked as intended. I also learned that designing for flexibility and clearly documenting interactions is key to preventing misalignment, as seen in a missed search behavior that initially led to display issues.