Medtech organizations are poised on the edge of analytics-driven transformation. The number of data sources available to medtech organizations has increased exponentially over the last few years, creating options, questions and confusion. How can you sort through these options to determine which data you should pursue? And once you find it, how do you assess whether it’s worth the investment?


The decision on what data to acquire and use for analytics is an important one for a couple of reasons: First, the sophistication of analytics being pursued and the ROI at stake have increased significantly over last few years. Therefore, it’s important to evaluate whether your data is the right fit to produce the desired results. Second, acquiring and managing data requires significant investment beyond its purchase. It also requires ensuring the end-to-end management of data integrity, usage, security, compliance, etc.  


In an earlier post, my colleague Maria Kliatchko weighed in on the many data sources that are available in medtech. Here are some examples of how these data sources have been used for analytics applications, as well as some considerations around their usage.  

  1. Sales: This area traditionally has been driven mostly by reporting that uses limited data sources such as sales and activity data. These traditional (yet important) applications include performance tracking and resource planning.

    Some organizations have started to use newer data sources to generate more accurate and real-time insights with significant ROI. One global medtech company used data from HCP sales, conferences, inventory, rep visits and demographic info to better understand its customers and drive growth after a major acquisition. Through a data-based approach, the company ensured appropriate customer coverage and exceeded sales growth targets by 12%. Amazingly, all of the data that it leveraged was already available within the organization.

    Publicly available data sources, such as CMS data, allow for the development of better value propositions for the hospitals (often used by the key account management team). One diagnostics company used this data to clearly differentiate its offering on metrics that are top of mind for its customers and achieved an estimated 7% incremental revenue.
  2. Marketing: Marketing analytics have gained much more importance recently as organizations look to build their commercial models on customer needs and outcomes instead of product features, and to optimize promotional spending and drive competitive advantage.

    Customer segmentation can now incorporate a variety of attitudinal and behavioral metrics by leveraging claims data, device-generated data, consumer data, channel preferences, etc. These data sources are enabling organizations to shift toward dynamic targeting approaches and personalize customer engagement with the right message, the right timing and the right medium. We typically see a 2 to 4% increase in sales from the deployment of such solutions.
  3. Pricing and contracting: While a lot of organizations treat P&C as an area of significant importance, analytics in this area are behind the curve compared to other industries. Many of our clients, even the large ones, complain about a lack of analytic rigor, data and tools in P&C, and express a need to use competitive insights and accelerate decision-making. However, a few organizations are rapidly advancing. One company applies machine learning with its sales data, contracting data, customer data (hospitals, IDNs, GPOs) and affiliations data to find opportunities to bundle products and develop personalized pricing offers, replacing segment-based price windows.
  4. R&D/product development: Significant innovation is already happening from a product development standpoint, encouraged by the availability of real-world data (including device-generated data) and the FDA developing standards to use real-world evidence for clinical decision-making. Companies can now use device-generated data in a variety of creative, new ways to assess the health outcomes of device usage, pursue new indications and develop value-added services.

In this blog post, we have just scratched the surface on analytics applications for the many data sources that exist. While a general awareness helps, it’s important to use a systematic approach to assess the applicability of data for any analytics application.


After locating the right data, you’ll need to assess its suitability based on data quality, integration, and cost and delivery issues. First, a data quality investigation includes questions around coverage, granularity or depth, biases, specificity, etc. Data that is fit for one purpose may not be fit for another. As an example, open claims data has good breadth of coverage, whereas closed claims data has better longitudinal coverage of patients. The former may be more suitable for targeting, and the latter for adherence analysis.


Second, assessing for data integration means learning how readily you can link the data to other data sets to supplement its weak points. For example, some analytics applications require the integration of patient data (like patient service program data) with claims data (which is anonymized). This needs to happen in a HIPAA-compliant environment, ensuring that patient identity is never compromised.


Third, when investigating cost, also consider delivery. You’ll want to ask questions around turnaround time and refresh frequencies. Some public data sources often take significant time (several months) and approvals, and come with higher usage restrictions.


We see tremendous opportunities on the horizon, not only to make better decisions but also to drive business model innovation through data and analytics. Exciting times are ahead.