As a growing practice, carriers are building and improving Web sites intended
to provide agents with information and functionality – presumably to
make it easier for agents to do business with them. Though agents get benefit
they’re not altogether happy about the proliferation of carrier Web
sites. Typically, use of the sites interrupts standard agency workflow, may
result in double entry of data, necessitates yet more training for agency
staff, and requires dealing with yet another logon/security process.
The results of the AUGIE (ACORD User Groups Information Exchange) Agents Survey
(see (www.acord.org/augie/ and the ACORD Presentation Center link for an overview)
show clearly that agents believe that carrier Web sites have a serious negative
effect on their productivity. One agency management system user group, ASCNet,
has adopted a position explicitly putting agency management systems, not “proprietary
software” in the middle of agency workflow. Some agency supporters have
publicly referred to carrier Web sites as the Dark Side of insurance technology.
And agents do have reason to complain. The more carrier sites they have to
deal with the more problems they face. Agents have to contend with one (and
sometimes more) security arrangements for each site. That means managing an
increasing number of logons, passwords, and a variety of mandatory password
change cycles. Data the agency keys into its rating or management system often
has to be re-keyed into forms on the carrier Web site. Agency staff must break
out of standard, management system workflow and then navigate through the
vagaries of the appropriate carrier Web site to finish their tasks.
It’s not a pretty picture and it’s far from the ideal agents have
been told is right around the corner – via real-time interface, an XML-based
communication process that puts carrier systems in the background and agency
management systems in the foreground. So more than one agent wants to demand
of his carrier partners “Quit doing agency portal sites and do real-time
interface through my management system instead.”
Though it sometimes seems to agents that carriers are blind to their needs and are pursuing “proprietary” solutions (a.k.a. carrier Web sites) out of narrow minded, competitive ignorance, in fact carrier Web sites have the potential to do much more for agency productivity than classic SEMCI (upload and download batch interface) and even modern SEMCI (real-time conversations between agency and carrier systems). SEMCI presupposes a traditional division of labor between agencies and carriers, with agencies submitting business and carriers underwriting and issuing it. SEMCI sees straight-through processing as the ideal, that is, a more efficient version of the standard insurance multi-step process.
At least some carrier sites take a different tack. Rather than perfect the multi-step, traditional agency/carrier division of labor, they collapse the process into one step, controlled and completed by the agent. Rather than straight-through processing, the carrier sites provide one step processing, sometimes called “once-and-done” or “one-and-done.” It’s in carriers’ interests to implement once-and-done solutions because they have the potential to reduce expenses much more significantly than straight-through processing. It’s in agent’s interest to use once-and-done processes because then they can both exercise direct control and obviate the need for follow-up.
Of course, one could grant the value of once-and-done but insist that it take place via real-time interface, batch interface, or a combination of the two, but in any case within the confines of agency management systems. But though doing everything through a management system may be attractive on the surface, a closer look reveals problems – especially in the area of carrier control. The critical concept is that in order for carriers to support once-and-done they need to control the process and they can’t do that, in any practical way, if the process takes place in someone else’s software, for instance a comparative quoting system or a management system.
Some examples: Let’s say that the carrier wants to provide a rich reference resource that helps the agent appropriately classify the risk as part of the once-and-done process. How will that reference material show up in the management system or be supported by either batch or real-time interface? Or what if the carrier wants to exercise control over when MVRs are ordered – by case and by agent – rather than letting the agent decide? How would that work delegated to a management system? Or what if the carrier requires significant additional information not supported by the agency management system?
Theoretically, it may be possible to conceive of an ad hoc solution to every problem caused by forcing once-and-done into the confines of a management system, but that doesn’t mean it makes any sense to try to do it. Clever companies will continually modify their rating, underwriting, and issue processes – in ways that will not be supported by current management systems and interface processes. So isn’t it more practical to bring agents to the company once-and-done process rather than try to bring protean once-and-done processes to the agent?
By the way, one could suggest at this point that remote Web services are exactly the right approach to bring carrier functionality right into the agency management system environment. Problem solved? I don’t think so – at least not for the next several years. Positing remote Web services as the (imaginary) solution merely begs the question.
From my point of view, the most plausible way today to deliver once-and-done functionality is through carrier Web sites. So here’s where we are: Once-and-done is better that straight-through processing (SEMCI). Practically speaking the best way to provide it is through carrier Web sites. But carrier Web sites are incompatible with management system centered workflow. Aren’t we faced with an irresoluble dilemma?
Not quite. If we can find a way for agency management systems (and comparative rating systems) to co-operate with carrier Web sites, we may be able to provide both carriers and agents what they want and need. So the real, practical issue that should be under discussion in the industry is not whether carriers should have Web sites but how to achieve cooperation between agency technology and carrier Web sites. What I call Hybrid Interface is a reasonable candidate to solve the problem.
There are many ways we could conceive of agency management systems cooperating with carrier Web sites. I’d like to propose a very simple kind of Hybrid Interface that would provide agents significant benefit while requiring little development effort from carriers and vendors. Let’s call it Passive Hybrid Interface. Formal surveys, anecdotal evidence, and common sense suggest that agents have three major problems with carrier Web sites: security, navigation, and data entry. Passive Hybrid Interface is intended to address these three problems.
Here’s how it could work. Let’s say an agent has data about an insured, drivers, and vehicles stored in her management system and wants to complete the sale of an auto policy on a carrier Web site. The agent would go to that insured’s record in her management system (establish a context) and choose the appropriate carrier from a multi-choice link in the management system (establish a destination). Then in the background, the management system would dump out the context (in some standard format) and let the carrier’s tray program (memory resident) know. The carrier’s tray program would pick up the context data and then use it to signon (if necessary), navigate and pre-fill the Web form(s) on its own site– allowing the agent to continue her work process at exactly the right place on the carrier site. Once an agent has established context and identified the right carrier, the transfer process and positioning in the carrier Web site should take only a few seconds.
Certainly there are other and perhaps better ways to implement Passive Hybrid Interface. The point though is to see that it wouldn’t be much of a problem for carriers or vendors to make it happen or for agents to use it. The agent would be in charge of creating the right data and context to trigger the appropriate management system response. The management system (vendor) would be in charge of formatting and dumping out an appropriate record and invoking the carrier program. The carrier’s tray program (carrier) would be in charge of interacting with the carrier site. All in all, appropriate assignments for the parties involved. And by the way, this approach would certainly apply to inquiry transactions as well.
Is Passive Hybrid Interface practical? I think it is. Would it provide a measure of cooperation between agency technology and carrier Web sites and be helpful to carriers and agents? Absolutely. Though so far agents have been opposed to carrier Web sites, my guess is that they’d change their minds provided the carrier sites were well done, provided real benefit, and cooperated with the agents’ management and rating software. What do you think?