Finding more value from managed care data across an organization with Tamr

In a previous post, I wrote how life science companies can gain an edge on their competition by constructing a well-laid foundation of curated data for each entity in the very complex managed care ecosystem. While this may not be a novel insight, Tamr is one of a few scalable enterprise solutions out there that can tackle the challenges of creating a well-adopted reporting layer from multiple data sources and data providers. Tamr’s unique machine learning approach of data mastering at scale, significantly helps reduce resources needed in implementing new data sources and maintaining the inflow of new data over time.

In this post, I will dive deeper into this idea, and give examples that show how implementing a best-of-breed solution like Tamr to curate your commercial data will greatly improve the efficiencies and impact on operational and strategic analytics throughout an organization.

Identifying the primary users of managed care data

Who are the data consumers relying on clean and curated commercial data? Typically, the end user of managed care data will be a part of one of four groups. Each of these groups bring different requirements, and in many cases are using the data for different purposes. However, what’s universal is that in order to be useful, this data needs to be  clean, accurate, and curated. Below are the four typical users within a life sciences commercial team:

Sales teams

  • The sales organization needs an understanding of the managed care ecosystem in order to effectively deliver and communicate value to their customers, who are often the physicians prescribing the company’s medical drugs and products.
  • Sales also need easy access to answers for critical questions like “what are my physician targets by segment and geography?” or “what reimbursement concerns are being voiced by physicians?”
  • Lastly, curated entities around managed care entities, formularies, and health plan profiles, for example, can help the sales team be more persuasive in driving adoption of the company’s medical products.

Marketing teams

  • Marketing teams are tasked with reaching  both the providers as well as the patients. To be effective, marketers need accurate data to best target these two audiences.
  • Marketers can also mine the data to answer key strategic questions, such as “what are prescription fulfillment rates across regions over time?” and “what are the best performing markets?”
  • Rapid insights supported by data from formulary, prescription, and claim sources can enable the team to optimize their marketing strategy whether it is direct to consumer, brand promotion, or couponing.

Managed market / market access teams

  • The managed market or market access teams’ analysis, based on managed care data, is critical to the company’s bottom line. The accuracy and relevance of their financial models are only as good as the data they depend on.
  • By answering questions such as “what are the prices and discounts we can provide to a payer in different regions?” or “what are the reasons for certain prescription reversal or rejection?”, this group is essential to negotiating the best contractual terms for the company in relation to other managed care entities in the market.
  • With the managed market team, accurate, curated data that can be produced quickly is even more critical as small percentages of error or delays in analysis can translate into millions of dollars of impact to revenue.

Health economic and population health teams

  • While not directly driving bottom-line impact, the health economic or population health teams rely on well-curated data to better understand and forecast the healthcare environment where the  company is involved
  • The work done by these teams are instrumental to informing the company’s longer term strategy from R&D to sales and marketing.

How to design well-curated data for managed care entities

Life science and healthcare companies have used Tamr’s machine learning approach to combine and consolidate key entity data, extracted from the variety of data sources they have procured. Instead of having analysts and subject matter experts wrangle disparate datasets every time there is a business question to solve, Tamr machine learning enables continuous curation of standardized data for key entities without manual or programmatic data wrangling. This approach frees up data analysts and data experts to focus on more pressing and bigger impact work than data prep and data maintenance.

Moreover, Tamr’s approach greatly enhances the overall data quality for reporting analytics. The example below illustrates a data quality issue where provider data and payer data communicates different information around relationships within the Intermountain Healthcare organization. Because of hierarchical and nomenclature differences in the data provided, trying to run an analysis that involves provider-payer may run into challenges around consistency and relationships needing to be reconciled by a subject matter expert. Other analysis done with these entities will consistently encounter the same problems, and lead to inconsistent outcomes.

Tamr’s approach to reconciling data sources with different nomenclatures and hierarchy is to  link all varieties of the same entity into a master record. This not only simplifies the entity relationships within an organization like Intermountain Healthcare, but also serves to provide standardized nomenclature – which will result in consistent, error-free analysis downstream.



The goal of Tamr’s implementation is to build a scalable and sustainable reporting infrastructure that allows data users to traverse the relationships of different managed care entities easily, without having to worry about data source quality.

As depicted in the illustration below, having a well-curated reporting layer for key entities greatly accelerates the process in solving key business questions.



Using Tamr as the MDM at scale

For many reasons, Tamr’s agile, machine learning approach to data curation is one of the only ways to scale for tackling the volume and variety involved in data sources capturing managed care data. While there are a variety of ways to implement Tamr in an organization’s dataops infrastructure, a proven strategy is to place Tamr within an organization’s data lake to perform data curation on staged data from disparate sources, and pipe the unified data into data management and reporting layers as shown below.

Considering the complex relationships involved in managed care data, leveraging a system that takes a machine learning approach to curate entity data significantly reduces the total cost of ownership for a system limited by writing and updating rules.



Compared to monolithic legacy systems, Tamr operates heavily on well-documented API services, and thus can easily weave into any infrastructure configuration as needed.


Many core business functions rely on a well-laid foundation of curated managed care data in order to effectively carry out the data analysis needed for business success. By using Tamr, customers have been able to finally realize control and maintenance over their data assets crucial to performing key data analysis, and only use a fraction of the costs and effort compared to traditional approaches. To learn more about the role of Tamr in the Life Science commercial business, reach out or schedule a demo.