THE TRUE COST OF NOT HAVING ENOUGH STAFF

There are businesses out there right now that think they’re saving $200,000, when they’re actually costing themselves $2 million.

Yet the solution is right in front of them.

The impact of insufficiently resourcing a financial services operation is felt acutely in insurance claims departments, where the staff are not only the processors but also the strategists and decision makers.

The amount it costs to employ these staff members is far outweighed by the big-dollar decisions they make every day. It’s why having a scientific approach to resource planning is essential – and not having one could prove very expensive.

Getting resourcing right for any processing area is an important management responsibility.

Inadequate resources lead to either poor service for customers (through delays and backlogs) or excessive overtime and pressure for staff, leading to poor culture and attrition.

Both are unsustainable.  

If areas are over-resourced, productivity suffers, costs increase and complacency occurs, making it difficult to return to full productivity when required.

Again, this is unsustainable for most operations and getting the balance right is a challenge.

WHY THIS IS IMPORTANT FOR INSURANCE CLAIMS TEAMS

For teams that manage complex portfolios such as insurance claims teams, the impacts can be even more significant. Losing control of a portfolio results in more than just delayed customer service, it leads to ballooning portfolio sizes and, often, increased duration in claims.

As the portfolio size grows so does the apparent resource shortage, and the issue can escalate quickly. The result? A hit to the bottom line.

A shortage of one or two full-time employees on $100,000 per year can easily result in millions of extra dollars spent on claims reserves.

Of course getting the balance right is requires a good model for determining the workload and proactive resource management behaviours.

What is different, and far more complex, is the nature of the model to calculate the workload.

Here, I’ll outline poss-e’s claims model that has proven to be effective across all claim types and mixes of business.

FIRST, IT’S WORTH NOTING THIS

As the goal of any resourcing model is to ensure adequate resourcing not just now but into the future, an accurate estimate of the future workload is paramount.  How workloads are forecast is highly dependent on the nature of the business and is a subject of its own.

The focus of this article is turning a forecast in to a resourcing plan and, as such, we’ll take it as given that a forecast of the raw work (in volumes of transactions, activities, claims) is being provided as an input.

CONSTRAINTS-BASED MODELLING

A resource model should be driven by the constraint that generates the workload.  In most cases these are pretty simple to determine.

Take a simple example of an agent that processes address change requests in a bank. A typical method to determine the workload is to set a standard unit of time to process a transaction and then multiply by the volume expected.

For example, an address change takes an average of two minutes to process and I get 60 requests per day - this is the equivalent of two hours of work per day. Once I have expressed my workload in a unit of time – hours in this case - I can convert to resources (full-time employees).

Of course, there is a lot more to factor in such as efficiency, leave, and other tasks. All this can be factored in by a good model, though the raw workload of unit time x volume is the foundation of an effective resource model. 

Let’s call this a ‘productivity-constrained model’, as I assume my only constraint is the time taken to complete all requests.  For a large proportion of back office operations, this simple model is enough to create a useful resourcing plan.

Another common constraint is the service level in a contact centre. In such environments, not only is the team constrained by the number of calls that need to be handled, they must allow for a service goal such as - 80 per cent of calls must be answered within 20 seconds.

The fact that calls can arrive simultaneously may mean that despite having sufficient resources to compete the workload, calls may arrive when agents are busy on other calls. A buffer may be required. 

For such situations, the maths of queuing theory is applied to allow for the impact of the call arrival pattern. It’s not uncommon for a contact centre to require 30 per cent more resource than is dictated by the volume of calls received. This fact is often ignored, though, which is why you sometimes sit listening to hold music for half an hour when you want to chat to your favourite call centre.

If we can identify what constraints exist in order to complete the work, we can build a reliable model.

THE CONSTRAINTS ON CLAIMS TEAMS

If we look at a typical claims team and try to determine the constraints, there are three that are easily identifiable.

Most claims teams recognise that effectively managing multiple claims at the same time creates a constraint. No matter the amount of touch time per case, at some point, staying on top of all cases becomes troublesome.  Hence the most common model is the ‘claims per person’ model.  Even when advertising for claims managers, it’s common for the job ad to highlight a typical case load of 40 claims per person.  This is one clear constraint for claims portfolios.

Another is related to throughput.  Assessing a new claim can take anywhere from 10 minutes (a simple life claim) to five hours (a complex IP or TPD claims). The nature of this work is strategic and there is a strong correlation between the effectiveness of this initial assessment activity and the eventual decision time. It’s a crucial phase of a claim. 

There appears to be a limit on the number of new assessments a manager can be expected to complete in a given time period. A typical example of this is two to three new claims per week, or 10 claims a month for a TPD assessor.

Finally, there is a productivity constraint. If claims processing is broken down into logical steps, such as ‘Initial Assessment’, ‘Subsequent Assessment’, ‘Claim Payment’, etc, it is possible to convert the volume of activity into workload hours. 

Using a claims forecast, a business can calculate the resources required for each of these three constraints. 

LET’S LOOK AT THIS EXAMPLE

A claims team receives 100 claims per month and carries an open portfolio of 1,000 claims at any given time.  The management team has agreed the following constraints and raw processing effort per claim is approximately 15 hours:

Constraint 1: Maximum of 40 open claims per case manager

Constraint 2: No more than five new clams per case manager per month

Constraint 3: After allowing for leave and other factors, the maximum output per case manager is five hours raw work per day.

Applying these constraints allows you to define the resource requirements:

Constraint 1:  1,000 claims / 40 claims per person  =  25

Constraint 2: 100 claims per month / 5 claims per person = 20

Constraint 3: 100 claims per month / 20 days = 5 claims per day  x 15 hours per claim / 5 hours raw work per person =  15

So we now have three different resource requirements - one for each constraint.  As each constraint must be met, we are bound to accept the highest value, which in this case is 25 managers, constrained by the claims per person rule. 

This example is typical of IP teams, who tend to carry large open portfolios.

For teams that manage life claims, the portfolios are much smaller, and the typical constraint is the productivity limit. TPD teams often bounce between a new claims per person constraint in some months, and open claims in others.  Our model caters for this, as long as we calculate every constraint every month, and select the highest number.

WHAT NOW?

A logical and transparent resourcing model like the one outlined above allows both management and claims teams to feel confident that the portfolio can be managed effectively.

 If the resource levels don’t feel aligned to the workload, understanding which constraint is driving the resource should be investigated and the constraint can be discussed.  Debating the appropriate constraints is generally a much healthier discussion than simply saying “we need more people”.

Forecasting and resource modelling are key features built in to the poss-e web application.  We want to provide managers with information they can understand and to be able to act on it. Leave the calculation and the reporting to the system. 

Continual discussion and debate on resourcing levels indicates a lack of confidence in how the operation is being managed. 

Getting resourcing right is vital to being able effectively manage your portfolio, so invest in the best tools to help you.

To learn more about how poss-e can improve your resource modelling, visit poss-e.com or send me a direct message on LinkedIn.