Dealer's Choice

While many insurers may hear the phrase generic process and wouldn't believe it's specific for their line of business, Kellington disagrees. "You may have a process that basically says, 'check eligibility rules,' and depending on what type of product you are writing, that service then can do something–either getting a medical history, doing MVRs, or whatever it is," he says. "The interface for passing in certain information and getting certain information back could be consistent."

However, Tana Sabatino, founder of Vallue Consulting and a former employee of ACORD, doesn't view the business process initiative in the same manner as ACORD. "Strategically it's nice to have a business process road map and identify the transactions that fall into place, but individual insurers have their own business processes, and I think where standards have the greatest impact is in solving the tactical problems of moving data back and forth between trading partners," she contends.

Membership participation and focus really is on the XML messages, Sabatino believes. "It's nice to have a road map, but at the end of the day, I need an XML message and I can't wait five years for that," she says. "If it continues on this path, I think [ACORD] is going to miss the boat where companies are going to have to go off and do their own thing."

Penn National is proceeding on its own with respect to an industry data model for property/casualty insurers. Bill Jenkins, CIO at Penn National, understands and values the business process models, but he just would prefer ACORD join him in trying to build the data model sooner rather than later.

Sabatino sees value in business processes, too–she's just not sure why they have to be published as a standard when carriers already have their own processes in place. "If you look at how much money companies have spent to improve their business processes, this is not going to help them because [carriers] still have to look at how they are doing things internally, with or without a road map from ACORD," she says. "And that's all it is–a road map."

IBM's contribution to ACORD represents a significant amount of intellectual capital in naming and describing processes for the industry, Kellington asserts. "Given that it's the entire business of insurance, the processes are at a high level, so there's quite a bit of work to do to drive toward an industry standard," he says. "It's important to say at this point the understanding of the big picture probably is more [significant] than the details of any particular process. It's difficult to work on the details of any particular process without a road map. IBM's contribution framed out the opportunity for us."

Some people are confused by ACORD's plan to proceed down the path of developing an industrywide process model, not a workflow model, Kellington continues. "Although our businesses are each unique, at some point in a workflow, there is a need to 'obtain a list of products for sale' or 'create a new party' or 'check eligibility rules.' These will happen throughout any process whether it's the creation of a life policy, a P&C policy, or even in the reinsurance marketplace," he says.

The workflow each carrier implements is unique and is not in the scope of ACORD's plan, Kellington explains. "It's beneficial to standardize these processes because they typically are consistent across lines of business, with the exception of product-related information," he says. "If we can standardize [processes], then the hope of creating a standardized SOA is achievable, which helps all of us in the end."

ACORD was looking at expanding the standards prior to the IBM contribution to include business processes, Garth claims. "It just so happened IBM offered up this contribution to help us jump-start this effort," she says. Process at a high level is not a competitive advantage, she points out, rather it's how businesses use that process within the workflow. "Taking some best practices from other industries and applying them to the insurance industry could be a real opportunity for the industry," she says.

Jenkins has worked closely with ACORD throughout his career and participated in the work done to implement ACORD's strategic plan. After the plan was put into place, Jenkins says ACORD asked him to canvas industry CIOs to determine what kinds of services they would like ACORD to provide. "ACORD has its own challenges relative to getting things to market quickly enough that people will adopt these things," says Jenkins. As part of his survey, Jenkins relates the general feeling of those he spoke with revolved around the development of an enterprise data warehouse. "What [the CIOs] indicated was if ACORD had a data model fashioned along the lines of the message standard, it would be a big help to them because it would allow carriers to talk more easily not only inter-enterprise but intra-enterprise," he says.

When joining Penn National a little more than a year ago, Jenkins found the carrier in the midst of developing a data model for an enterprise data warehouse initiative. "I reached out to ACORD and had it come up and give training so the people here understood what the ACORD messaging standard looked like and made sure we were capturing all the components and attributes required to align with the ACORD messaging standard," he says.

Jenkins also asked ACORD whether it would be interested in developing a data model for the P&C industry. ACORD has a data model on the life side, which Jenkins describes as a hybrid between a data dictionary and a modular data model.

ACORD decided to create one holistic data model that will consist of all lines of business–life and non-life–for all geographic regions, Jenkins says. The data model will be developed after work on business processes and a data dictionary is complete.

Garth doesn't see that as a delay and feels the data model is dependent on the common business dictionary and the business context of the business processes. "ACORD's strategy is based upon an architectural approach that is driven by the insurance business' needs: process and a common vocabulary, which drive the requirements and corresponding development of the data model," she says.

The foundation of a data dictionary is essential to building a business data model, Garth believes, since the data model is an abstraction of the dictionary. "It is a logical progression for optimal results," she says.

Jenkins decided Penn National needs to go in its own direction, based on the carrier's business needs. "[ACORD] set a time line for the end of 2008 to have a data model put into place," he says.

Following the work on business processes and the data dictionary, ACORD plans to develop the data model and look at messages, particularly ones that could be common, regardless of line of business, according to Garth. "We want to bring that all together so companies have some flexibility to do things in the manner that is most appropriate for their organization and look at things on an enterprise basis from a data perspective," she says. "At the end of the day, the data is the most important thing about the organization that really reflects the value of the organization and also reflects information as to decisions that have to be made to position that organization competitively."

If a comparison is made of the ACORD XML for Life vs. what is available in property/casualty insurance, Sabatino contends the life insurance industry has a data model as the basis of its standard, and all XML messages are based directly off that data model. Data is available regardless of the message. "So, I don't have to be concerned while I'm doing a quote I don't have the address I need because it wasn't added to the message," she says. "The information is in exactly the same place regardless of the message. The party address is the party address."

Ninety percent of the focus on the life data standard is on the data model, Sabatino explains, and the messages automatically can fall out of that. The standard also provides a basis for carriers to use internally because they have a full data model they can apply to their own internal transactions as opposed to just the transactions ACORD happened to define, she adds. "It's a key part of the life insurance model," says Sabatino. "You see a significant uptake on that standard for internal implementation and integration as well as external. You don't see that same level on the property/casualty side because [it doesn't] start out with an industry data model for property/casualty. It's message by message."

The life standard doesn't have the strength and flexibility needed to meet the internal integration problems property/casualty insurers face, according to Sabatino. "[P&C] insurers really are seeing the gap, and it's becoming more obvious because we don't have that gap on the life insurance side," she says.

For P&C, Sabatino maintains, the immediate problem is the lack of a data model. "The end result of that is huge in that you cannot take that [life] standard and apply it in a practical sense for internal sharing of data," she says. "It also poses limitations when you take the standard and use it for something ACORD has not yet defined. You have all those things defined somewhere, but it's not presented as a data model, and you can't use it for another business purpose."

Establishing a data dictionary is another important step, Kellington suggests. "Any meeting you go into, the first thing you do when you get a group of people together to go over some kind of flow or message is to get a commonality of what you are talking about," he says. Most lines of insurance have their own dictionaries. There was no hesitancy to do one for the entire industry–it just was a matter of getting it done, states Kellington. "If we are going to be creating some enterprise process models, you almost have to start with a dictionary first so you know what you are talking about," he says.

Garth points out not all of ACORD's standards grew organically within the standards group. She adds the insurance industry was very silo oriented in the 1990s. "This gives us the opportunity, because of changes within the industry, to bring it all together into a single dictionary," she says. ACORD already has begun its efforts regarding the data dictionary and will publish an initial content release during the first half of this year. "Basically the ACORD staff will bring it all together, make a recommendation, and then go through the process with our members to validate, agree, disagree, or make changes as necessary," says Garth.

The data dictionary is a key part of the data model, Sabatino notes. The data model essentially is the structure the data dictionary would fall into. "The data model puts structure and relationships around the dictionary data," she says.

Jenkins believes if ACORD is able to meet its time line, it will be a big benefit for the entire insurance industry. "But it's a matter of delivering it," he says." It's the execution that's always a bugaboo. You can't hold up the business to get things done at the pace of a standards organization."

When carriers step back and think about what ACORD is doing, asserts Kellington, it makes total sense, but he acknowledges there is a learning curve as ACORD attempts to get people to understand the value of a process model. "Once those [issues] are broken down, I think it will be a lot easier," he says. "It's more of a communication issue."

Some insurance companies will be the leaders of this movement, Garth believes. "As always, when you have leaders and pioneers–whether it's in new product development or a new way to do claims–that kind of change forces the rest of the industry to look at what's being done and do something different to keep up with the competition," she says. "The industry is not going to change all at once. There are going to be leaders who will take the opportunity and help create that change."

ACORD's strategy is based on input from and the needs of its members, explains Garth, with the goal being to help enable the industry to run, grow, or transform. She defines run as tactical, while grow and transform equal strategic issues. "ACORD continues to support those companies with tactical delivery of standards, which we are doing today," she says. ACORD has a large number of working groups going on in all three domains and cross-domain initiatives as well as work on version 2 of the Property & Casualty/ Surety specification. "In fact, we have been able to respond to all of the tactical requests for working groups," she says.

ACORD also is implementing a strategy built upon what Garth calls a "sense and respond" foundation. "Companies want to grow and transform, and ACORD will be there to enable transformation vs. waiting for 'requests' for enhancements," she says. "As with all companies, there always is a balance of tactical and strategic so that you meet immediate needs but also are preparing for the future. Our members have requested and are actively involved in the evolution and expansion of the standards to meet their growing and future business and technical needs."

Nevertheless, several carriers have concerns about shifting the focus to business process models, according to Sabatino, because they see it as a drain on resources that should be dedicated to the tactical delivery of standards. "I'm totally committed to ACORD, and I left ACORD to make this stuff [standards] real," she says. "But we have immediate problems. Let's solve those. We can keep our eye on the big picture, but I don't need to spend my energy on that."