Location-based provider search engine

Case study

Location based provider search engine

Workers’ compensation is insurance that helps organizations take care of medical expenses and convalescence for injured employees, lost wages, and death benefits for employees and their family members. According to the Bureau of Labor Statistics, in 2018, 2.8 million non-fatal workplace injuries/illnesses and 6,147 fatal injuries/illnesses were reported by private sector organizations.

Companies need to comply with the requirements set by the state and opt for workers’ compensation insurance. Workers’ compensation legislation varies from state to state in the US. A major challenge faced by organizations is their inability to help employees identify the right healthcare provider in their vicinity.

Insurtech can help organizations navigate through the complexities of workers’ compensation programs. Insurance companies are now focused on improving operational efficiencies by identifying the best healthcare providers for their beneficiaries of the workers’ compensation across multiple specialties and locations.

Our client drives change in the commercial insurance markets with easy to use digital solutions. These solutions substantially reduce claim costs by foreseeing client needs and aligning them with the best possible resources.


The idea was to build an insurance providers’ search engine with geo mapping, based on the client’s scoring engine. Our client wanted:

  • The search engine to help customer service representatives or insurance agents identify the best-rated healthcare providers or doctors across specialties and remote locations for employees.
  • An analytics dashboard to provide insights on healthcare providers.

The expected outcome of these features is to identify the best provider for every claim and a decrease in the claim handling duration and costs.


The solution had to be based on the client’s scoring engine and delivered on a SaaS platform. Some of the key challenges include:

  • Processing huge volumes of data and storing billions of records
  • Lack of customer-level customization options to drive analytics
  • Low latency leading to poor performance in loading times


We proposed to come up with a provider search engine which enables users to perform location-based search for top-rated providers. To enable this type of search, we had to build the provider profiles, that comes from various types of data sources. The data source is a CSV file which can be refreshed on a quarterly basis. In addition, fresh data can be captured from the daily claims load to identify new providers from the claims prescription.

We designed the database to hold the captured field elements from the CSV file and perform elastic search on them. This is to make sure that our client’s customers can create their own exclusion list of providers along with their list of network and out of network providers.

Users had to access healthcare provider information within a radius of few miles around their location. So we decided to use Mapbox to provide geography-based map search. We also enabled location based clustering of customers on the map and included a geographical heatmap of the customers. The heatmap facilitates location-specific service offerings. For example, when an employee calls the helpline, a customer service representative helps in suggesting a top-rated doctor within a radius of 5, 10, or 25 miles from the customer location, based on incident or speciality.

We enabled users to rate the services of healthcare providers and provide a cumulative internal rating score. This in turn will guide users to suggest the best-rated healthcare providers to claimants.

The analytics dashboard was designed to provide insights about each healthcare provider and their history of cases. The eligible search parameters included zip code, city/locality, provider name, provider specialty, NPI, Tax ID, and much more.

Tech stack

How Our Solution Helped

Scalable engine capable of searching billions of records in less than 60 seconds. Identification of top-rated doctors led to 1/4th reduction in claim costs.

Overall approach

We came up with a multi-tenancy architecture driven solution. The architecture helped to separate customer data, enable high-level customization, and master dataset versioning. Elastic search was used to process large volumes of data and present the results through the UI.


  • Improved performance of the application with quick loading of pages
  • Quick identification of top-rated doctors as per the requirements of the claim
  • Intuitive and user-friendly dashboards and charts
  • HIPAA compliant data security and privacy

Talk to us