Getting download right

Rough Notes, Aug 2003 by Anderson, Steve

It should work. Why doesn't it?

The Interface Requirements Report, first published in 1990, set the agenda for what we've come to know as company download. The report explained that uploading information from the agency's management system to the company was going to be too difficult a task to accomplish in a reasonable period of time. The difficulty stemmed from the need to include company edits on the agency system. A company edit process allows the information that is input by agency staff into the management system to be validated as accurate and within the carrier guidelines prior to being placed in the carrier's system. Because of the complexity of upload and the limitations of the technology at the time, developers of the report determined that download (moving information from the company system to the agency system) was where resources should be directed initially.

Here we are 13 years later and the question remains: Does download work? Or maybe more accurately: How well does download work?

Why ask those questions? When we talk with agents across the country as we're doing presentations, we hear over and over again that "download doesn't work" or it doesn't work well. Some agents are so frustrated with the existing process they're thinking about turning off download for some of the companies they represent. This becomes even a greater issue as more companies begin to turn off the agent copy of policyholder documents, thus requiring the agency to rely on the downloaded information to answer policyholder questions, or requiring them to rely on the carrier Web site to access accurate policy information.

Please don't get us wrong-download is a good thing. Anything that will reduce the amount of multiple entry that agency staff has to do should be implemented and used. So why would agents turn off download? Let me list several of the issues.

Overwriting data: One of the first problems-and one that seems to be the most prevalent-is the overwriting of data by the download received from the company. If the agency uses its management system to gather and store more information about the client and/or policy than is sent by the company, that data is either lost or moved to a note or memo area, where it's harder to retrieve. This isn't just a carrier problem. Many agencies sabotage their databases themselves by putting information in fields where it doesn't belong. This happens most of the time because there isn't anyplace to put a particular piece of information in the database. And when the download comes from the carrier and does use (or could use) that field, anything already there is overwritten.

All caps: The carrier computer systems are generally older legacy systems. These systems often store data in all capital letters. So, when this information is downloaded, it comes as all caps. This makes it difficult for the agency to use the data in this format, especially name and address information. Some agencies are re-keying this information into upper and lower case so they can use it in correspondence.

There is activity underway to improve this. The Workflow Automation Services (WAS) subcommittee of AUGIE (ACORD-User Groups Information Exchange) has recommended that support for mixed case download be publicized at upcoming AUGIE meetings as well as with carriers and management system vendors. This is something that vendors and carriers will have to work on implementing outside of ACORD.

Not enough data: In many cases download works well, i.e., data comes from the carrier into the right fields and in the proper format to the agency system. Even in this scenario, there can be a problem because the download doesn't provide all of the information that the agency needs for a particular policy or group of policies. There are frequently holes in rating or underwriting information, for example. This creates a problem for agencies because they don't have all of the information they need to answer all of the questions that might arise on the account. Without complete rating and/or underwriting information, the agency is unable to re-market the policy in subsequent years. The solution is for the agency to manually enter the missing data.

Time delay: The batch store and forward communication process currently in existence creates a time delay in getting the information back into the agency system. This creates a workflow issue for the agency. Suppose you enter information into the carrier system. Can you wait for 24 to 72 hours for the data to be processed and downloaded back into your system? Clients generally need an auto ID card or Certificate of Insurance processed as soon as the coverage is put in force. In cases such as this, the agency will need to enter the data manually because the download doesn't arrive in a timely manner.

So what's the solution? Potentially there are several solutions to this problem. First, carriers and vendors should make sure the existing download process works as well as it can. It does seem that after all of this time and effort the industry should be able to do a better job. Another solution may be the proliferation of some of the new, real-time communication processes between agency systems and carriers. Currently, most of these processes are inquiry only, but the trend is toward building interactive transaction processing also. Following is a brief overview of the some of these new services.