HealthSherpa Plan Recommendation

HealthSherpa is the largest private marketplace for health insurance under the Affordable Care Act (ACA), where navigating insurance complexities, particularly with the ACA, can be daunting. To simplify this process, I developed the Plan Recommendation feature to offer users an intuitive starting point focused on overall affordability.

My Role

I was the sole designer on this project. I also served as a product manager, given we did not have this function at HealthSherpa during this project.

Team

I collaborated closely with a lead engineer and a data scientist to identify solutions for the challenges encountered in plan selection.

Resources

I leveraged data analytics, user interviews, session data, and more to help inform the design solutions.

The problem

Choosing health insurance involves balancing financial, healthcare, and risk tolerance factors. It is especially challenging for those with low financial and insurance literacy.

We conducted hundreds of interviews, surveyed thousands, and analyzed data from millions of shopping sessions. This extensive research led to numerous enhancements in the plan shopping experience. The idea I proposed was a plan recommendation feature.

Who did I design for?

Consumers coming directly to HealthSherpa.com to find health coverage.

Non-benefits-eligible consumers offered health coverage support by their employer

Consumers specifically searching for plans from a particular health insurance company.

Where did I start?

After conducting numerous user interviews, surveys, industry research, and data analysis, I developed wireframe solutions to enable users to start the plan shopping process confidently and positively. Key insights included:

  1. Difficulty in comprehending the various financial aspects.

  2. Challenges in understanding industry-specific terminology.

  3. Overwhelm caused by an abundance of choices.

  4. The monthly premium was considered the most critical factor.

  5. Deductibles and out-of-pocket expenses were also highly significant.

Recommends the top three plans

Recommends the most affordable and lowest premium plan

What did we build?

The most effective approach turned out to be recommending a single plan. Our research indicated that presenting multiple options initially heightened the cognitive load for consumers. Starting with one anchor plan proved the best way to initiate plan shopping.


The feature set included:

  • Healthcare utilization selection

  • The most affordable plan view

  • Affordability breakdown (estimated all-in)

  • Estimated all-in costs listed on every plan card

  • Sort by most affordable

  • Healthcare utilization edit functionality

Challenges

  • The core shopping experience's technology was integrated into a shared code base. Whenever a carrier requested a unique feature in the shopping experience, we had to evaluate its feasibility and potential impact on both partners and consumers. Maintaining this balance was crucial and always driven by data and its anticipated effects.

  • User interviews, research, and testing revealed that consumers often had concerns about how their data, particularly related to plans and prices, would be utilized. This skepticism stems from pre-ACA experiences with underwriting and unfavorable practices by insurance carriers.

    In response, we prioritized transparency in the healthcare utilization step and the recommended plan view, striving to provide the clearest information possible to users.

The experiment on the recommended plan yielded significant results: approximately 60% of users chose the recommended plan, leading to a 10% increase in overall plan shopping conversions.