As stated in July, the more things change the more they stay the same - but how are systems checked and how are they assessed? Gill Salton explains.
The underwriting engine
The underwriter has now taken on a different form. No longer do they turn up Monday to Friday and work a seven-hour day sitting at the same desk, looking at a set number of cases. A large number are now an underwriting system/rules engine making decisions on up to 85% of the business being received at any time.
Underwriting systems have developed since their introduction in the early 1990s from:
- Static systems with non-interactive questions, where the answers were entered into system at the head office or branch, without the customer being present, to
- Customer-present fully interactive underwriting systems, asking drill-down questions which allow final decisions (standard rates, ratings, exclusions, declines) to be issued at the point of sale on a vast majority of cases.
Some systems are also built into the tele-underwriting process where the tele-interviewer can input the answers in real time and ask further interactive questions, or input the details once the interview has been completed.
Underwriting systems: potential for good
Underwriting systems have certainly revolutionised the way in which applications are underwritten. However, there needs to be a balance between accepting as many applications as possible and not compromising the claims position. There are various benefits both for the customer, distributor and from a provider/risk management perspective:
Customer / distributor benefits:
- Immediate acceptance for the customer.
- Ratings can also be accepted at point of sale.
- Saves time for the distributor.
- Significantly reduces the application to policy issue time.
- The management information available can allow comparison with other distributors with regards the business quality.
- Reduced claims issues as the right cases are accepted on the right terms.
- Overall, a good customer journey.
Provider / risk management benefits:
- Vast amounts of management information can be produced for analysis - straight-through processing rates, top 25 disclosures, NTU rates, etc.
- Provides more consistent decisions.
- Capacity to underwrite far more cases than their human equivalent.
- Peaks and troughs in volumes are easier to deal with.
Underwriting systems: potential for bad
However, with any system there is the potential for things to go wrong. As Bill Vaughan (American columnist and author) said: “To err is human, to really foul things up requires a computer.”
With the high number of cases being ‘accepted’ each day by a system, an error in the questions being asked or the decisions being applied can affect a far greater number of policies. In addition, these errors could continue unnoticed for a significant period of time.
Obviously, the more ‘intelligent’ the system the more chance of an error occurring and the higher the impact. Important questions to ask are:
- Who will pick this up?
- How many cases will it effect?
- What will the cost be?
