A “member enrollment” challenge in group employee benefits that is far deep-rooted than was thought to be, and how a “next-gen” solution can fix it.
Why member enrollment is an area of concern!
Member enrollment is a crucial business function in healthcare and insurance group benefits and is directly associated with customer experience. It influences the experiences of all major external personas in the following way:
- Eligible employees start receiving coverage protection only when successfully enrolled in the carrier’s system
- Employers depend on the successful completion of the enrollment process for determining premium installments and accounting
- Plan administrators get paid only when the enrollment process is complete
- Brokers receive commission only on successful completion of the enrollment process as the way to determine employee participation
What are the key challenges?
From a group insurance customer’s (or employer’s) standpoint, the member enrollment process is time-consuming that span weeks or even months. Such a lengthy process creates a poor enrollment experience for employers and their employees.
From a carrier’s standpoint, enrollment is a high resource and cost-intensive process. The cost of enrollment per member is very high. Such a high cost of enrollment impacts the carrier’s bottom line.
What lies at the root of these challenges?
We must see how the enrollment process works for most of the group and health insurance carriers to get to the root of the issue.
Member enrollment traditionally has the following scenarios as below:
- Enrollment in bulk that takes place systematically through the upload of census data mostly for large employer groups (2000+ lives) in a “True Group and List-Billed” benefit plan; multiple members (potentially in thousands) get enrolled all at the same time through this process systematically.
- Online enrollment for individual members, where one member gets enrolled at a time through manual or systematic data entry, through an enrollment portal (or a Benefits Enrollment Platform, like BenefitFocus, ADP, etc.), often as self-service or assisted by the Plan Administrator(PA). This is normally the common practice for small groups who enter into agreements with the carrier for voluntary benefits for their employees.
- EOI (“Evidence of Insurability”) or special medical underwriting at the member level, which is a common business requirement for both bulk and individual enrollment in the below scenarios:
- When a member asks for coverage exceeding the employer-stipulated ‘guaranteed issue’ coverage limit (or GI limit)
- When a member seeks enrollment outside the employer’s prescribed auto-enrollment or Guaranteed Issue (GI) window
Out of the above three enrollment scenarios, bulk enrollment and EOI have more pain points than the online individual enrollment. Why?
This is because Bulk enrollment and EOI are two functions that each carrier defines and architects based on their unique business perspectives and requirements (including data model, business rules, etc.). So these functions significantly vary across carriers from a functional standpoint. This makes a generalization of these functions difficult from a technical implementation standpoint.
That is the reason why NOT many tools or solutions are available commercially for these functions. Carriers traditionally implement these two functions by developing custom solutions in-house and deploy those solutions on-premise (often due to lack of “cloud” skills, and other preconceived notions (for example, data security)).
That’s where the pain points for these two functions originate because the maintenance of these custom apps is expensive and resource-intensive, and also offer limited capabilities from business operations and customer experience standpoints.
In contrast, Individual member enrollment is way more matured, generalized and vanilla in the industry. Many Benefit Enrollment Platforms (often available as SaaS or BPaaS offerings) are available in the market that meets the needs of most carriers from a benefits enrollment standpoint. Also, individual member enrollment as a capability is often outsourced to TPAs as a valid option.
Thereby we will focus on the Bulk enrollment and EOI as the themes for this study.
What are the challenges in bulk enrollment!
If we take a look at the bulk enrollment process in the context of “True Group & List billed” plans, the precursor to this process involves receiving census data (usually in large volumes) from different employers by the insurance carrier. This is achieved through the integration of partner systems (carrier and employer) largely leveraging SFTP protocols.
Traditionally each employer sends out its census data to the carrier in its unique layout and format (comma-separated, tab-separated, JSON, XML, etc.). It is traditionally on the carrier to decipher, translate and standardize these incoming files into a common template or standard for ingesting, storing and eventually processing the census data for enrollment.
Most carriers do not follow any industry-standard message model for the intake and ingestion of census data that they receive from different employers. They attempt to invent their own (canonical) messaging model. This makes data standardization a complex operation for the carriers that often introduce “quality” issues in the system.
What more, the carriers normally deploy large-scale ETL ecosystems, that are built and deployed on-premise, for receiving incoming files of different shapes and sizes (from different employers) and standardize that incoming data for ingestion and storage using their self-invented canonical message models.
Such ETL ecosystems need regular servicing and enhancements for acquiring new employers who come with new data standards, layouts, and formats.
These ETL ecosystems need constant servicing for existing employers as well when these employers make changes to their existing data standards, layouts or formats for census and eligibility feeds that are sent out to the carrier.
Such regular maintenance and enhancements to these ETL and data assets to address needs for new and existing customers make it an insanely complex, IT-intensive and costly operation for insurers.
Additionally, these ETL jobs are extremely compute-intensive and traditionally run as long-running batch processes due to the need to process humongous volumes of the census and eligibility data on-premise. Such a batch-intensive mode of the operation increases latency and reduces the availability of information that has poor implications on customer experience.
Can industry-standard EDI-834 files help to standardize the format?
One can argue that standard formats are available in the industry in the form of (EDI) 834 files in which employers can communicate their employees’ health insurance enrollment and maintenance data to insurance carriers. However, the truth is that EDI 834 file format has not been able to make the process sufficiently simpler. The reason is that although the format itself is standard meaning all record types and properties are classified in the same way, the information contained within the properties can differ from carrier to carrier. Meaning, if you have one file configured for UnitedHealth, that same file can’t be sent to Aetna, Humana, or really any other carrier because the structure of the data is fundamentally different for different employers.
Furthermore, once you expand the EDI 834 layout beyond conveying more than just health insurance data to a carrier, to include something like dental plan data, carriers will typically require a different format.
Can data quality be a challenge too?
Yes, it sure can. It makes things worse when files coming in from different employers contain corrupted or non-standard data values, nulls or outliers. Fixing such data quality issues require manual data stewardship which is a resource, cost, and time-intensive exercise.
So where do we land with all of these!
It can be easily concluded that all of these that are discussed above create a number of critical challenges for the carrier:
- High Capex in assembling a complex on-premise ETL ecosystem for the intake of census data from different customers/ employers
- A non-standard data model for intake; most carriers try to re-invent the wheel by creating a message model of their own, for data standardization, without adopting an industry-standard
- High OpEx due to constant IT servicing to the ETL systems to acquire new customers, meet the changing needs for existing customers, and manage the quality of data; High OpEx leads to poor ROI
- High latency and poor availability of information lead to poor customer experience
What are the challenges in the Evidence of Insurability (EOI) or Medical underwriting
EOI or “Evidence of Insurability” is a group benefits provision that states that a member (or an employee) has to undergo special underwriting process, regardless bulk or individual enrollment, if any of the following scenarios holds good-
Enrollment outside Guaranteed Issue (GI) window – Member who needs to be enrolled outside the stipulated time window for automatic enrollment (or Guaranteed Issue / GI period). Example – employees who have joined employment after the employer’s prescribed auto-enrollment (or Guaranteed Issue) period is over ate subject to EOI check.
Coverage requested beyond GI coverage limit – Member who requests coverage amount that exceeds the employer’s maximum allowable coverage limit for Guaranteed Issue (or GI). Example – If an associate asks for 4000 USD monthly as income benefit for LTD payout while the maximum allowable guaranteed monthly LTD payout for an associate in his/her designation as stipulated by the employer is only 2000 USD then the associates is subject to the EOI check.
With the above understanding let’s see how EOI processing works and what are the challenges!
The EOI process is traditionally a manual underwriting workflow process for most carriers. Eligibility and coverage details for members who qualify for EOI (or Individual Medical Underwriting, as this is also called) are routed or directed to human underwriters for manual underwriting. This process is inefficient due to a number of reasons:
The underwriting software, that these human underwriters use, normally support a very limited number of underwriting rules; they are normally standalone applications with very limited straight-through capabilities. Most of these applications have grown outdated and are not capable to handle most of the underwriting scenarios in the context of the modern group insurance business.
Due to such limited capabilities, these underwriting systems require a lot of IT development & maintenance work to keep those systems relevant and up-to-date. Such frequent IT servicing adds to OPEX costs and also increases system downtime and eventually non-availability of business capabilities.
Lack of systematic straight-through capabilities leads to a significant amount of manual administration for underwriters to arrive at eligibility decisions. This increases the turnaround time of the overall enrollment process that impacts customer experience, and also adds to costs of manual work (or Opex).
BUT WAIT! The industry has a large number of modern Benefit Enrollment Platforms like ADP, BenefitFocus, Rippling, GoCo, etc. which provide best-in-class digital engagement to employees and plan administrators for enrolling members to group benefits. Why don’t these platforms solve the problems that we articulated above?
Because benefit enrollment platforms, despite being modern and rich with digital engagement features (like self-service, omnichannel, etc.) do not address the issues that we discussed above.
The problems that we have highlighted are back-end problems which exist in the layer that sits in between the customer-facing benefit enrollment platform and carrier’s core systems of records. In other words, the problems in our focus area lie in the ingestion and processing of the benefit enrollment data that are received from the benefit enrollment platforms or from the employers via SFTP. Those problems do not necessarily exist in the customer-facing digital engagement layer.
What’s more, modern benefit enrollment platforms traditionally address the needs of the online enrollments at the individual level. They normally do not address the needs of bulk enrollments and EOIs which are the major focus areas of this study.
Thus modern “digital” benefit enrollment platforms are not sufficient to solve the whole spectrum of “enrollment” issues. While those platforms do solve a part of the problem mainly from a customer engagement standpoint, they can be best described as “lipstick on an ugly pig” where the ugly pig is the back-end process layer that hides behind a “glamorous” benefit enrollment platform.
Which part of the enrollment ecosystem we are solving for?
The below schematic diagram illustrates where the problems are that we need to solve for in a member enrollment ecosystem, and a summarized list of those problems.
- Poor enrollment experience due to high turnaround time
- The high cost of enrollment due to high IT servicing needs and manual administration
- Poor utilization of underwriting resources
- Poor management of quality of the incoming census data
A real-life experience:
It was a gorgeous day in Portland Maine when I was chatting with my long-time pal, Surrendran, who was blessed with a baby boy six months back. The proud father indeed looked a happy man, but soon when our conversation rested on annual enrollment to workplace benefits the smile on his face seemed to fade away.
“It was a horrible experience this year when I enrolled me and my family to our benefits at work. It was as painful as it can be”. Surrendran looked stressed out and clearly upset when he spoke out those words.
I was little taken aback and inquired what went wrong! I knew from one of my prior conversations with him that his company uses BenefitSolver enrollment platform to enroll employees and their families, which I thought to be a modern enrollment platform with cool features that allows self-service & multi-channel capabilities. I was not sure what could make his enrollment experience so poor! So I asked Surrendran what the matter was. What I learnt from him was pretty amazing and insightful.
Surrendran said that his company has recently changed group insurance carrier for providing employee benefits. So this year the enrollment happened with the new carrier. While Surrendran has no complains about BenefitSolver, he truly believes that the new carrier has a way to go in “fixing” their systems to be able to make the experience any better.
Surrendran’s enrollment took around 6 months to be completed, and initially he had no clue what could take it so long! After several rounds of calling, the Benefits team told him that the “batch process” of the insurance carrier was unavailable because it was going through an IT enhancement to serve his company who is sending census data in a “different format” that the carrier’s systems do not currently support.
Also there are a lot of structural “issues” in the data that his company sends that the carrier’s IT team is fixing manually.
What’s more, the carrier’s EOI capability is largely manual and it often takes months for the carrier to complete the EOI process for just one member.
How do we describe Surrendran’s enrollment experience! “Nightmere-ish!!” There are millions out there who are passing through the same experience as Surrendran at the time of their annual enrollment.
On the other hand, if we see the carrier’s side of things, the carrier’s bottom line is hit big-time for every new employer being acquired and added for employee enrollment due to the high costs resulting from significant IT development effort, high infrastructure costs, significant manual administrative overhead, and cost of error. A poor bottom-line impacts the carrier’s stock performances.
How can we leverage technology to help the carrier and its customers get rid of these issues! That’s where our “solution” story begins.
What is the solution?
Before we get to the solution, we will summarize the problem statements as below:
What makes the member enrollment complex for Group Benefits?
- Ever-increasing benefit admin platforms with a growing employer base
- Ever-increasing volume, velocity & variety of incoming enrollment data
- Need for carrier’s resources (people, products & infrastructure) to scale and adapt quickly
- Poor customer experience and high cost of enrollment on failure to scale and adapt efficiently
How do we manage this complexity?
What are the key questions that need to be addressed?
What are the questions that need answers in the bigger context?
- How will the carrier ingest and standardize member enrollment data when received from multiple and disparate benefit admin platforms?
- How will the carrier manage the quality of incoming member enrollment data?
- How will the carrier simplify underwriting in member enrollment?
- How does the carrier apply their specific enrollment rules that are not supported by commercial-off-the-shelf benefit admin platforms?
- How will the carrier utilize its historical data to bring in more efficiency in risk management and data quality management?
- How do resources auto-scale on-demand to manage an ever-growing volume, velocity, and variety of enrollment data?
What is the recommendation?
The solution that I conceptualized, designed and prototyped is a cloud-native SaaS offering that solves poor customer experience and high cost of member enrollment in group benefits for healthcare and insurance industries.
It provides an on-demand, real-time, straight-through and integrated enrollment experience at a substantially low cost with AWS and Machine Learning as the enablers.
The platform is founded on the IntelliDataSmart platform which is my other innovation that provides a scalable, resilient, secured, cost-effective, serverless, data processing pipeline that enables flexible intake of data in any format & schema, and automated metadata management, configurable data standardization, and automated data quality management leveraging AI/ML. The platform can be leveraged for a wide variety of data-centric use cases including enrollment, claims, data lake, and analytics.
The solution got placed in the final four spots out of 321 ideas submitted in a global competition for the Next Generation of Insurance technology.
Here is my certificate of recognition:
