#FutureReadyHealthcare
Machine learning-powered recommendation engines
How Indegene is helping global life sciences companies win the personalization battle
The Customer: The story of a rare disease drugmaker
In the competitive realm of enhancing healthcare professional (HCP) experiences, leading global life sciences companies are prioritizing tailored omnichannel engagement strategies.
One such company was on a similar mission. In order to increase prescriptions for its drug designed to treat a rare kidney disorder, the company aimed to gain profound insights into the HCPs operating in this field. Their objective was to better understand the preferred communication channels of these HCPs, their favored types of content, and the timing they preferred for engagement.
Armed with these valuable insights, the company wanted to empower its sales representatives with personalized recommendations, enabling them to establish more impactful connections with HCPs.
Like sending Dr. Jane Doe an invitation to a webinar related to treatments for Abnormal Hemolytic- Uremic Syndrome (a rare kidney disorder) via SMS, her preferred channel for engagement
With data-driven approaches like these, the company hoped to elevate engagements and foster better connections with their target HCPs.
The Challenge
To personalize engagements for its HCPs and boost prescriptions for its kidney disorder drug, the company needed to proactively drive better engagements with the right content at the right time. However, accessing data insights to make this possible posed a challenge. Additionally, the lack of a solid, explainable recommendation engine set them back in terms of providing representatives with reliable and highly contextual recommendations for every HCP engagement.
What they needed to succeed
What they needed was a recommendation engine that was intelligently built to:
Provide highly contextual recommendations for every HCP engagement
Ensure that these recommendations were not only backed by data but also transparent in logic
Enable representatives to proactively drive better engagements with the right content at the right time
Provide a seamless HCP experience by integrating personal and non-personal engagement channels
Uncover untapped opportunities by identifying potentially valuable HCPs that are currently lower on the priority list
The Solution: Three-Phase Strategy Employed by Indegene to Address the Company's Needs
Our strategy included three broad phases:
1. Discovery
We extensively engaged with sales, marketing, and analytics teams to evaluate brand readiness.
CONTENT
We analyzed HCP segments
CONTENT
Explored available data sources
CONTENT
Assessed the quality, format, and frequency of datasets provided by vendors
CONTENT
Analyzed how the company’s sales force is currently structured and mapped to existing promotional strategies
2. Planning and design
We developed a comprehensive roadmap for building, operating, and expanding the capabilities of the decision engine
We considered strategies to increase adoption rates as well, including change management. For the initial phase, we recommended introducing the engine to a only subset (e.g., 50%) of the sales force
We incorporated an Explainable AI (XAI) model to provide transparent insights into the decision-making process and thereby instill a higher level of confidence in the recommendations provided by the engine
3. Develop and Deploy
Created an analytics-ready dataset (ARD) that included data like:
CONTENT
Patient comorbidities
CONTENT
Payer plan details
CONTENT
Prescription fills over two years
CONTENT
HCP referrals and community clusters
CONTENT
HCP activity across personal and non-personal channels
Transformed ARDs into Model-Ready Datasets
CONTENT
Aggregated features at the HCP level
CONTENT
Target variable: new prescriptions
Training of the XAI model using ensemble techniques
The model was trained with the objective of providing recommendations that are focused on HCP engagement in a way that leads to an increase in new prescriptions
CONTENT
We selected the best-performing model based on test dataset accuracy
CONTENT
SHAP values were utilized to infer key features
Implemented rigorous testing methodology prior to finalizing the model
CONTENT
Implemented semi-automated testing
Optimized recommendations across personal and non-personal channels
Leveraged:
CONTENT
HCPs’ channel preferences
CONTENT
Field intelligence
CONTENT
Sales rep feedback
Disseminated recommendations to downstream teams
CONTENT
Seamless integration into the Veeva CRM system
Established standardized governance practice
Outcomes: Achieved a 10-15% increase in new HCP leads for brand targeting
With Indegene's solution, the customer successfully attracted fresh HCP leads for targeted brand initiatives, witnessed increased adoption rates among the company's sales representatives for decision engine recommendations, and experienced a notable reduction in the manual effort required for validating these recommendations.
10-15%
Increase in new HCP leads for brand targeting
SLOT
70%
adoption rate among company’s sales reps for decision engine
recommendations
SLOT
15-20%
reduction in manual effort required
for validating recommendations
SLOT
Supported by
10-15%
SLOT
10-15%
SLOT
10-15%
SLOT
10-15%
SLOT
10-15%
SLOT
10-15%
SLOT
Ready to transform your HCP engagement strategy? Contact us today to learn how our machine learning-powered recommendation engine can drive meaningful results for your brand