4 mins read
02
.
12
.
2020

How to win more business by prioritising broker submissions

By Andrea Camacho, Product Manager, Cytora

Prioritising submissions will instantly increase your premium, profit and client lifetime value, according to Cytora’s Andrea Camacho

Your underwriting teams are currently leaving over 20% of premium on the table, and you probably don’t even know it.

At Cytora, we’ve interviewed a large number of underwriters who have told us that there is no consistency in how incoming submissions are prioritised today. Even across underwriters working in the same team. Some underwriters focus on the most urgent submissions first, some focus on the ones they think will require the least effort, or maybe those coming from the most insisting brokers. Very often underwriters simply work on submissions on a first in first out basis (FIFO).

While on a high level the concept seems very simple – work on the most valuable submissions first – executing effective prioritisation is really challenging. There are two main reasons for this:

  1. The value of a submission is unknown. Underwriters can try to guess which ones are good risks or bad risks, but to do that they need to spend time analysing the submission.
  2. Not all submissions convert. In fact, just a few are ultimately bound. Predicting which ones is a daunting task.

These two problems result in what we called the Prioritisation Paradox:

To prioritise submissions, underwriters need complete information. To get complete information they need to analyse every submission. This defeats the very purpose of prioritisation, which is to decide where to invest your time, before you invest your time.

With all this in mind, it’s no surprise that underwriters struggle to prioritise their work and resort to largely unsuccessful strategies. However, there are now ways to solve this.

Early adopters are starting to adopt predictive technologies in insurance to support their underwriters. The most successful ones are following these steps:

  1. Extraction: key risk identifiers are automatically extracted from incoming submissions
  2. Profiling: the risks are augmented with hundreds of data points sourced externally to build a digital profile of the risk
  3. Prediction: machine learning models run through the data to predict the risks’ value and chances of conversion
  4. Prioritisation: submissions are prioritised and surfaced to underwriters to maximise the expected value of the portfolio

But hold on! We’ve talked about generating more value by prioritising submissions. How much are we really talking?

At Cytora, we’ve modelled this to quantitatively measure the value for the commercial mid-market. For the purpose of this analysis, we’ve defined value as lifetime value of the risk – net present value of a policy cash-flow during its lifetime. The conclusions can be generalised for other metrics such as risk premium or profit.

Working with a number of insurers, we’ve concluded that:

1) A submission’s lifetime value follows a curve well described by normal distribution, with a positive mean but a large standard deviation that makes a large percentage of a policy’s lifetime value negative.

In simple words, while the majority of clients generate a profit, a large portion generate losses.

2) There is a strong negative correlation between submission conversion rate and quote turnaround time – defined as the time that an underwriter takes to analyse a submission and return a quote to a broker.

This sounds reasonable, the longer you take to get back to a broker the less likely you’ll win the business. But interestingly, our analysis has shown that the correlation is best described by a logistic curve that starts from a base conversion rate and after a period (which mainly depends on the line of business) abruptly decreases until tapering down to zero.

Keeping these two points in mind, we then simulated the value that could be obtained if underwriters were able to process submissions from the highest to the lowest value and compared this to the traditional FIFO approach.

To do this we used typical distributions of underwriting processing time and submission receival time. We also assumed a team of underwriters working at capacity.

We’ve run the model for 500 simulated hours for an individual (approximately 1 month of their time). We’ve demonstrated a theoretical increase in the portfolio lifetime value of over 20%.

While the uplift may be higher or lower depending on an insurer’s specific factors, there is little doubt that prioritising submissions is a key strategy to put in place in underwriting teams. It will avoid you leaving value on the table and will instantly increase your premium, profit and client lifetime value.

To find out more about the wider augment, filter and prioritisation process, take a look at our previous blog here.