Maveryx Success Stories

Learn how Alteryx customers transform their organizations using data and analytics.

Showcase your achievements in the Maveryx Community by submitting a Success Story now!


Propensity to Pay

10 - Fireball
Name: Charity Wilson
Title: Senior Consultant
Company: Saxony Partners
Saxony Partners logo.png
Overview of Use Case

Cashflow is a major issue for physicians as they have little insight into how much an insurance company will pay them for a visit. They feel as if they are working endlessly, but rarely know the amount of income they will receive from one month to the next. We helped a physician group gain better insights into their income using linear regression in Alteryx. This Alteryx workflow provides the necessary visibility required from the physician’s group, increased morale amongst physicians, and allowed the group to make more strategic decisions based on data.


Describe the business challenge or problem you needed to solve  

In the Health Care Industry, a physician has no idea how much income they will receive from month to month. This is even more evident among small physician groups, where cash flow is often a complete mystery.  Their cash flow is based on contracts with the insurance companies. The insurance companies dictate what the reimbursement rate will be and to put it simply, the physicians don’t have the time to spend reading through the contracts. Contracts tend to be long, confusing, and include a number of nested statements. Many times, physicians sign their contract without a full understanding of the reimbursement policies. Physicians typically have 30-50 of these contracts being processed at the same time and it is nearly impossible to keep track of what will be coming in from one month to the next. Government and legal policies are updated, causing a domino effect to insurance reimbursement rates.  As a result, an individual physician can feel like their cash flow resembles a black box. They submit invoices to the insurance company, and don’t know what they’ll get back.


Describe your working solution
I created an Alteryx workflow to help a physician group gain visibility. To create this model, I first had to understand how insurance reimbursements fluctuated historically. We looked back at three months' worth of historical data on insurance reimbursement rates to physicians to train the model. Using linear regression, I then created a statistical model of the reimbursement rate.  I looked at a variety of things, such as diagnosis code, and if the patient paid a co-pay.  By focusing on just the past three months of data, I could hone in on the current cash flow situation.  As policy changes come in, this model is adapting to them in real-time. Once the model is refreshed each month, I can then forecast the insurance reimbursement rate for each of the patients who was seen.  The forecast is 88% accurate. On average the forecast is within $1 of the actual insurance reimbursement rate.
Describe the benefits you have achieved
The physician group we are working with is so excited. The insight from Alteryx is much better than anything they have ever had. They didn't have any type of forecasting models to begin with, so the information coming from the new workflow is extremely helpful. So helpful, in fact, that our app dev team is working on getting this workflow embedded in an app that we are designing for the physician.  So, benefits are still yet to be accrued in full.  
Before Alteryx, physicians had no way of knowing what their monthly payments would be. In order to accurately predict reimbursement, there must be near real-time updates of the model with all insurance policy changes. We created an algorithm that looks at historical data to give them a forecast that is accurate and it is dynamically responding to the changes that are happening in real time. 
To put it simply, Alteryx will allow each physician to actually know how much they will get paid. It makes it less risky to go ahead and hire that next person, or invest the money in that next capital investment. For example, the physician group that we're working with is getting ready to acquire a new physician.  With the historical data, I can tell them how much cash flow the new business will bringThe biggest value expected will be to new physicians just out of medical school.  Right now, new physicians feel like they are working their butts off, and have no clue if it's even worthwhile.  This will help with team morale.