What happens when a large transactional system that is scoped to manage around 2,000 applications a month suddenly faces an unprecedented influx of 2,000,000 applications? You quickly discover that your legacy system needs help.
Recently, I had an opportunity to sit down and talk with Dr. Matthew McCarville, former Chief Data Officer for the state of Florida, about how Florida responded when faced with an unprecedented influx of requests for unemployment assistance at the height of the Covid pandemic.
In the face of the pandemic, Florida’s unemployment agency found themselves facing an overwhelming challenge – the inability to scale with the sheer volume of new applicants. In a matter of days, a task force of state technology and data leaders worked to supplement their existing system with critical automation to handle the extreme influx of applications. Using a combination of Google Forms for data intake, analytics automation through Alteryx, and RPA through UiPath, the agency was able to restore access to a critical array of services for those seeking unemployment relief.
Two million applications and counting
The pandemic caused a sharp influx of unemployment applications in Florida (from 2,000 a month to 2,000,000), and their unemployment system was unable to scale in response to the requests for support. That led to a system that was inaccessible to those desperately seeking assistance, and the information that was collected could not be analyzed for eligibility, adjudication decisions could not be processed in a timely manner, and manual processes could not ensure adequate protection from instances of fraud. This led to delays in the distribution of unemployment relief, creating additional economic pressure on individuals and families, as well as pressure on state leaders to swiftly correct these concerns.
Agility and stability through analytics automation
By implementing a Google Forms-based process, the team was able to fix the problem associated with the input of application information. Once the data was collected, the Alteryx Analytic Process Automation (APA) Platform was utilized to process and analyze this data and create an integrated dataset with other sources to automate the process of identification (fuzzy matching), the determination of eligibility, and the management of funding distribution. Additional workflows were automated, including the use of Alteryx to feed RPA processes (UiPath), to handle the adjudication of claims.
The core system was still limited to 150,000 concurrent sessions, meaning that on average, 400,000 users needed to have their input (data collection) handled through Google-based processes. Through an integrated process that included both APA and RPA, analytics automation was applied through 72 decision tree algorithms to handle backdated claims and provide unemployment relief automatically and retroactive to the date of filing instead of when the application was eventually processed.
Building for the future
With a combination of technologies and the application of analytics automation through Alteryx, Florida’s unemployment agency was able to recover from the increase in applications, restore a critical element of the state’s social safety net, and ensure that people received the unemployment support they were entitled to faster and more efficiently. Additionally, the state is better equipped to handle future challenges, and is benefiting from a higher level of automation that is making operations more responsive and efficient.
Watch This Next.
You can watch a discussion with Dr. McCarville recorded during the 2021 Alteryx Inspire event.
Read This Next
Leveraging Alteryx for a Successful Job Search
Learn how Mike Davis' experience with Alteryx put him ahead of the competition.
2021 Alteryx Podcast Wrap-up
In this special crossover podcast episode, hosts Maddie Johannsen and Susan Currie Sivek relive some of their favorite standout moments and special guests.
What Does the Democratization of Analytics Look Like for Government?
For organizations like the US Navy and others, the ability to better leverage data will be built on the capability to upskill the domain experts in the use of analytics.