May 2020

Challenge

One of the largest affinity-based insurance carriers, with over $2.6 million active policies split over several disparate applications, was handicapped by the multitude of systems of record and countless data inconsistencies. The company was facing customer data acquisition, audit and compliance issues. There was a strong need to improve the customer experience, reduce the financial impact of discounts, cross-sell and up-sell more effectively.

Solution

Saama Analytics devised a strategy and maturity roadmap with the carrier, and built a Customer 360 solution for a single, consolidated view of customers.

• Better tracking of critical KPIs than just mapping planned vs. actual stages
• Significant saving of resources and cost
• Ability to predict and correct anomalies
• Increased probability of keeping the study within budget and maintain timelines
• Advanced AI capability to enable better decision making

The Customer 360 solution was built on these components

1. Master Data Management
2. Data Quality services
3. Enterprise Data services
4. Customer portal

Overview

Before any drug or treatment received regulatory approval, it goes through rigorous testing and development. While determining the efficacy of a drug is paramount, patient safety cannot be sacrificed to achieve it. Patient safety is  a crucial part of clinical trials and has traditionally been the responsibility of medical monitors. Saama Analytics’s Life Science Analytics Cloud (LSAC) Drug Efficacy and Patient Safety analytics capability allows the monitoring of patient adverse events, lab data, vitals, and concomitant drug interactions.

When you look at multiple sites and disparate data systems, it becomes clear that critical information about the way a particular group of patients is responding can become lost or obscured in the mass of data.

LSAC, as an AI-powered solution, has the potential to automate any data process and leverage it to derive critical insights that may be the difference between successfully executing patient safety protocols and suddenly encountering a high-risk situation during a trial.

Benefits

A crucial part of setting up a clinical trial is establishing various controlled groups to determine a drug’s efficacy. The process is quite sensitive and has many built-in safety practices. An AI-powered solution can help ensure that patient safety data is more accurate and has the ability to warn medical monitors of potential adverse events or risk.

While medical monitors evaluate all processes and ensure patient safety, they could use some help in sifting through data and identifying patterns not immediately visible to the human eye. An AI-powered solution can perform a deep dive into any data system, harmonize it to adhere to the Industry standards – Study Data Tabulation Model (SDTM) – and then extract insights quickly and accurately. These insights may have been otherwise unidentifiable, and even if a group of analysts had been able to draw the same conclusions, they would not have been able to do so as quickly.

LSAC can ensure protocol adherence, keep data records standardized and free of error, monitor inclusion/exclusion criteria, keep a constant check on all regulations, and generate near real-time analysis of data. With real-time insights at your fingertips, it is possible to have a comprehensive view of how a drug or treatment is performing across patient pools and sites on a global level. Since clinical data is standardized near-real time using a metadata based approach, it allows sponsors to standardize the clinical data management, review, and monitoring process. This approach enables the submission pathway to be faster and more efficient.

LSAC also offers a virtual assistant interface that provides access to results from the Drug Efficacy and Patient Safety Analytics capability. As a result, activities such as the review of clinical data, troubleshooting sites with a high number of adverse events, and performing patient-level safety analysis, become more accurate and efficient.

Business Problem

Creating a winning deal for customers requires in-depth thought process and planning. Herein, the Brand Managers of our client found themselves in a tough spot, answering tough questions like…

“Should I run this promotion with retailer Walmart first or Costco?”

“Should I offer a higher discount on single units or design a multi-buy promotion?”

“Should I run the promotion in the 1st week or 2nd week of Feb and for how many weeks?”

“What if… I increase the budget by 10%? Will I get higher sales? How it will Impact my ROI?”

Questions kept coming, and the planning process increasingly became complex. And due to the sheer volume of products and tight deadlines, Brand managers simply tweaked their earlier strategy. They had shifted approximately 35% of the last year’s promotion plans with a minor change in discount.

The impact? Profitability suffered because the team wasn’t able to achieve optimum performance with their planning approach.

The SpendO Way

SPENDO stepped in with a solution to transform promotion planning with AI. Our platform helped the company to –

• Learn from historical data patterns identified and communicated by AI
• Simulate millions of promotion scenarios
• Optimize every single promotion for better results
• Prescribe a winning deal

Earlier, the team could manually run few What-IF scenarios before selecting a promotion deal which they were not sure would give back optimal results. SPENDO ran 1000 different What-if scenarios for every single promotion, on the click of a button.
All the brand manager had to do was provide:

• The product category,
• The quarter they are trying to promote the brand,
• The retailers targeted for this promotion

SpendO provided optimized end to end promotional plan with details like which retailer, which product, starting the week, ending week, how much should be invested as Trade Spend and what will be the outcome if this promotion plan was to be executed.

Brand Managers could also compare the optimized plan with last year to see YOY improvement in results. Accordingly, they now had the knowledge to increase/decrease the budget for specific products and measure the outcomes.

The Challenge

A large Insurance carrier was facing a number of challenges with claims management and fraud detection.

First, the company had amassed a significant amount of claims data through the years. However, the data was difficult to consolidate and use. The information resided in five different legacy claims applications, and the insurer had no enterprise-wide view of claims.

Second, the insurer’s claims team was unable to perform any real time analysis. A claim is considered ‘the moment of truth’ for policyholders. Churn and dissatisfaction are directly correlated with an individual’s experience with how expediently their claim is processed and settled. It was, therefore, essential to optimize the claims process.

Third, there was a dire need to automate existing fraud models and build system that would provide effective fraud prevention techniques to mitigate premium leakage.

The Solution

The Insurer used Insurance Analytics Cloud, which is phased, agile, and iterative. With this method, the company was able to realize ROI in shorter cycles, rather than at the very end. A phased approach also ensured a smooth transition, and allowed for data accuracy validation on the go. A glass claims conversion pilot was performed and later built upon.

“Prior to this project, we were only able to refresh a small amount of claims data. Now we have more data, and we can see it in real time. We can see things as they are happening. we can’t act on it as fast as we are finding it.”