Everyone loves the idea of analytics making their lives easier, but in practice medtech companies don’t always realize the broader goals of their analytics programs. Even when analytics initiatives are well planned and executed, they often fail to yield their expected returns due to one factor that often slips under the radar: the human element.
Many medtech companies embark on their analytics journey with a focus on building their data management and technology infrastructure capabilities. For example, several medtech companies have invested in some sort of technology solution to improve data access, reporting and analytics production. Others have gone one step further and set up analytics capability groups to incubate advanced analytics skills within the organization. These investments run into the millions of dollars. However, there’s little planning or investment in ensuring that the end users will take the required actions based on the analytical insights that these investments generate.
End users of analytics often suspect insights that go against intuition. This can be exacerbated by factors such as not knowing what to do with the insights in the real world and inertia in taking action when the personal incentive isn’t clear. Although the exact roadblocks may differ depending on each organization’s realities, some of the critical hurdles that most companies face include getting end users to believe in the analytical insights, getting them to understand how to act based on these insights, and enabling them to appropriately walk the talk.
Addressing these hurdles requires a three-pronged strategy:
1. Pick your battles. Many organizations approach these hurdles through organization-wide, big bang initiatives, which creates resistance to change. It’s imperative to pick a few key areas within the organization where stakeholders are convinced of the value of analytically driven decisions. Prioritizing which areas to pick requires an understanding of three things:
- The magnitude of change if the analytical insights were to be acted upon
- How the stakeholders and end users in that area might react to the change
- Whether the incremental value driven by this change warrants the efforts to embark on this journey
We’ve been on a journey with a medtech company since early 2017 in which one of the pressing business problems included inaccurate demand estimations that led to both shortages in supply and inadequate deployment of commercial resources. To address the situation, the company decided to develop machine-learning-based forecasting models that led to a 50% decrease in prediction error. This success story was heavily and deliberately publicized throughout the organization with a focus on the resulting impact rather than the “cool” machine learning models behind the scenes. This led to stakeholders across the organization realizing the tangible value that analytics could add to their current roles and thus generated advocacy across the organization.
2. Minimize the time spent deducing insights and actions. The ability to deduce your own insights from data on a self-serve analytics workbench may sound like an exciting prospect to analytics practitioners. While this approach has its benefits, like democratization of analytics and more control over the insights by providing end users the ability to tinker with the underlying assumptions, it’s imperative to understand the end user’s motivations and skill profile before embarking on the self-serve route. For example, the sales rep’s primary motivation is to sell, and any action suggested by analytical insights that helps them sell better will receive immediate buy-in. However, a “cool” self-serve analytics workbench that expects the sales reps to play around with the data to generate their own insights and deduce the actions will be met with resistance. For sales reps, generating insights isn’t enough. The action to be taken based on the insights also needs to be made explicit for end users.
3. Enable the action by overcoming conscious or unconscious objections. In theory, getting stakeholders to believe the analytical insights and understand the resulting action may seem like enough to drive the action, but it rarely suffices. There may be a multitude of inherently human objections that cause the inertia: the personal incentives of taking the action may not be clear to the end users, or the action may require the end user to step out of the user’s comfort zone, which can carry a perception of risk. We believe that the key to overcoming these challenges is to develop a deep understanding of the end users’ concerns regarding why they might not want to act—even if they believe the insights—and then develop a mitigation strategy. This is not trivial and will require strategic investment and leadership support.
For example, a medtech company implemented sophisticated analytical models to address challenges with customer service. Its customers constantly complained that the services offered were not timely and did not meet their needs. This had started to show in market share declines. The analytical model took the suspense out by recommending the optimal sequence of touch points and interventions that would help the company offer the right service at the right time. Despite this, the company continued to lag in NPS scores. On further introspection, the company realized that the challenge wasn’t with the quality of the insights but with getting the sales force to act according to the insights, suspending their own beliefs and methods. In many situations like this one, addressing biases proactively is what’s required to go the distance.
Even the most groundbreaking insights from the latest analytical model won’t drive incremental revenue or cost savings if no action is taken based on those insights. It’s just as important to understand the challenges posed by human factors and implement capabilities that address these challenges as it is to develop the next cutting-edge advanced analytics models. Medtech companies that address the human and technology elements will be poised to realize the full potential of their analytics investments.