Every insurance company could benefit from further streamlining their claims process. P&C insurers often spend up to 80% of total premiums on claims processing and payout. Traditional approaches to determining the “as-is” state of business processes are a time-consuming, costly, and not always accurate process. The promise of process mining is that it removes the guesswork from how actions are being executed by accessing the data from the involved systems.
Five critical steps companies need to take to use process mining to optimize the claims process:
Identify the value opportunities to improve, and the KPIs and metrics to track
Measure how the claims process performs
Identify and prioritize the root causes for performance gaps
Identify and prioritize the solutions to eliminate the performance gaps
Implement the solutions, measure the improvements, and tune
Step 1: Identify Value Opportunities, KPIs, and Metrics
The macro value streams to increase the optimize the claims process are:
Reduce direct operating costs (e.g., for adjustors, car rentals, incoming calls)
Reduce supporting operating costs (e.g., IT infrastructure, HR, finance, and other administrative functions related to claims processing)
Reduce excess claims payout
Increase or at least maintain customer retention
Ensure cost-effective compliance
Which of these five would provide the most value to the claims process? For example, focus on not just reducing operating costs (1 and 2) but overall claims loss (1, 2, and 3). Once the opportunities are identified to pursue, that will determine which KPIs and metrics you should track.
For example, here are global KPIs that help identify the root causes impacting value opportunities:
Claim settlement cycle time, measuring the average amount of time to settle a claim, from (first notice of loss) FNOL to formal settlement, can be used to measure performance and problems in loss (1, 2, and 3 above) and retention – as long cycle times cause both leakage and customer dissatisfaction.
Cost per claim (measuring the total cost of processing claims) and claims loss (measuring the total amount of claims paid out plus any loss adjustment expenses) can be used to measure problems in loss (1, 2, and 3 above).
Going down a level of detail, these two more specific KPIs will help identify the root causes of performance problems:
Percentage of claims with supplements. Measuring the accuracy of the damage assessment of the adjusters, can be used to measure problems in all value opportunities.
New and closed claims per adjustor. Measuring the productivity of adjusters, can be used to diagnose direct and supporting costs, though closed claims per adjustor is also useful for diagnosing satisfaction and thus retention.
And finally, we find these even more specific KPIs and metrics to be most useful because they point most clearly at root causes and thus at the possible solutions to remediate them:
Cycle time between FNOL and initial customer contact after loss event
NIGO (not in good order) or claim error rate
Claim payment cycle time or other measurement against an SLA
Percentage of claims processed with digitization and automation
Customer satisfaction, as measured by NPS (Net Promoter Score)
Number of quality assurance activities, which are themselves inefficient and indicate problems in the process
Inbound call volume, which also are inefficient and indicate problems in the process
The next step is to use process mining to measure how your claims process is actually performing.
Step 2: Measure How Your Claims Process Performs
We’ve explained process mining for claims in several places, so for a more detailed explanation see here.
Process mining is a new AI-based technology that helps to quickly and effectively optimize complex business processes like claims management, in ways that were not possible just a few years ago:
It provides a data-driven way to analyze and optimize your processes, making process improvement more science than art.
It accelerates identifying the areas of the process to automate and optimize.
Process mining works by analyzing event logs and other data from claims management systems and workflow tools to identify process improvement and automation opportunities. Our clients typically first use process mining to discover the process friction points and the root causes of those inefficiencies. Typical friction points and their causes include gaps in process standardization, rework, and lack of automation.
By knowing the empirical root causes, companies can make targeted changes to the process to resolve the issues. And then after making the first round of enhancements, you can monitor and continuously improve the process.
Doculabs provides a Claims Execution Management Application built on the Celonis process mining platform to provide an end-to-end view of the claims process by extracting data from our client's source systems, such as Guidewire or Duck Creek. We can also extract event log data from document capture systems such as Kofax and automation platforms such as Appian and Pega.
The result of the process mining exercise is information on how you are doing in the KPIs and metrics you’re investigating, and – as you drill down – where you have gaps in performance, and the root causes of those gaps.
Step 3: Identify and Prioritize the Root Causes for Your Performance Gaps
There are three primary root causes in common across every business process performance gap:
People: inadequate training and staffing, inadequate organizational structure, lack of checklists and job aids
Process: lack of standardization and consolidation, too many exceptions and redundant steps
Technology: Lack of digitization and automation
Here are some examples of the typical causes of performance gaps that we find:
Claim settlement cycle time: Often caused by inadequate claims sorting procedures (with claims assigned to the wrong adjustors), poor inbound data quality (e.g., NIGO claims), and poor claims adjuster training. More specific cycle time performance gaps (e.g., from FNOL to first contact or from FNOL to repairs completed) are associated with more specific causes.
Claims cost: Often caused by inadequate claims sorting and routing procedures, low adjuster productivity, NIGO claims, or a high rate of rework.
New and closed claims per adjustor: Often caused by inadequate training and documented best practices for adjusters, poor claims assignment procedures for adjusters, or a high backlog of unopened claims (for new claims).
Step 4: Identify and Prioritize the Solutions to Eliminate the Performance Gaps
The following framework of solutions include:
Technology: Digitize and automate tasks, focusing on document and data ingestion, document and data understanding, communication anywhere it should happen, automating actions, and automating early QA or validation steps. Automate the orchestration of your tasks and processes using a low code workflow platform. Leverage intelligent capture, intelligent document processing, and RPA. Read more about these issues here.
Process: Redesign processes to ensure standardization, consolidation, reduce exceptions, leverage automation and real-time next best actions. Process mining has revolutionized process redesign and migration. Read more about these issues here.
People: Address personnel with training, checklists, early and automated QA, and job aids.
Specific levers to pull that impact KPIs include:
For cycle time issues:
Assign complex claims to more experienced processors and adjustors
Proactively communicate with customers throughout the settlement processs, using AI-based tools where possible to automate communication tasks
For cost-per-claim issues:
Consolidate claims process steps when possible
Define SLAs for how quickly adjusters perform inspections after FNOL
There are usually many candidate solutions that could be applied. So it’s important to prioritize the solutions and put them in a roadmap to implement.
Step 5: Implement the Solutions, Measure the Improvements, and Tune
In some ways, once you’re “finished” with the above, the real work begins, because in this step you implement the solutions to move the KPIs. However, by using process mining to uncover the root causes of process friction, you can make targeted changes to the process to resolve the issues. Then, after that first round of enhancements, keep going.
Use process mining to continuously monitor and manage the claims process. You can monitor not just historical but real-time performance by tracking KPI measures, identify process bottlenecks as they occur, and use the AI-based capabilities of process mining tools to get specific recommendations to increase performance levels.
A great first step to begin optimizing insurance processes with process mining is with Doculabs' help. To see how, watch our Claims Execution Management Solution demo below or get in touch with us via phone at (312) 433-7793.