| Syndicate Assignment | Manzana Insurance Fruitvale Branch | Operations | | | 01 June, 2016 | |

|

Manzana’s Position in the Insurance Industry
Manzana has a history of seeing opportunity and seizing it. Manzana has grown through acquisition. Manzana's main competition is Golden Gate Casualty. They beat Manzana on both price and performance. Golden Gate tied with Manzana for first place in the property insurance market in 1988. By 1991, Golden Gate announced they could guarantee turnaround time (TAT) of one day. If they fail to reach that benchmark, they offer a 10% premium discount. Manzana must at least match that promise to stay competitive in this market.
The current TAT for request processing has bloated to five or more days at a time. This report will recommend a number of strategies to improve the system in operational efficiencies, measurement and service. Manzana has an opportunity to increase their business through decreasing their TAT, adjusting staff incentives and delighting their customers.
The acquisition of Manzana by Banque du Soleil has helped. There was a restructure that resulted in a leaner company. Underwriting standards have tightened, operating expense has reduced and this has resulted in a regain of market share. After the restructure, they are now better geared toward the property liability insurance market. Manzana now specialise in commercial insurance rather than personal insurance. That has pitted them against Golden Gate. For commercial insurance, the revenue breakdown of Manzana is: Category | Proportion of Manzana Revenue | Property Insurance | 65% | Liability Insurance | 20% | Investment Income and Miscellaneous Specialty Lines | 15% |
Figure 1: Manzana revenue by product category
As illustrated in fig. 1, liability insurance is only 20% of Manzana's revenue. The liability insurance crisis of the 1980s saw huge payouts to customers. That meant reduced profitability and a low proportion of resources allocated. The Fruitvale branch of Manzana increased the proportion of resources working on property insurance. This was to avoid taking on the high risk of payouts that would threaten the branch's profitability and meant it could stay open.

Analysis of Manzana
In order to take on the challenge that Golden Gate has thrown down, Manzana must analyse their systems and performance. We will make some recommendations based on an analysis of Manzana’s * Key Performance Indicators – measurement and incentives * Cost structure * Bottlenecks

Assumptions
This analysis includes these key assumptions: 1. The exit time from the processing of a previous step becomes the arrival time of the next step, e.g. CVp (Distribution) = CVa (Underwriting)

2. The arrival variability at the entry is nil: Arrival of policies to the distribution appears to be perfectly random with no discernible patterns of peaking 3. To complete this analysis we have taken the mean and standard deviation from the Policy Processing Times by Department in 1986 4. See Figure 2 and 3 for assumptions based on Manzana’s performance Time Assumptions | Minutes in an hour | 60.0 | Hours in a working day | 7.5 | Working days in a week | 5.0 | Weeks in a month | 4.0 | Days in six months | 120.0 |
Figure 2, assumptions used in time-based calculations Resource Assumptions | Rating clerks | 4 | Distribution clerks | 8 | Writing clerks | 5 | Territory 1 underwriter | 1 | Territory 2 underwriter | 1 | Territory 3 underwriter | 1 | Pooled underwriters | 3 |
Figure 3, assumptions used in resource-based calculations
Key Performance Indicators (KPIs)
The KPIs for Manzana Fruitvale branch are benchmarked against targets for the current quarter, the current quarter vs. the previous year and the current quarter for Golden Gate. They are: * Number of new policies * Number of endorsements * Number of renewals * TAT (days) * Renewals late (%) * Renewals loss rate (%)
Management signal to staff what is important to them. At present, it is: TAT, the percentage of late renewals and the percentage of renewals lost. The announcement by Golden Gate of the one-day TAT and 10% discount if it is not met has increased the pressure on these KPIs.
The targets for Manzana's Fruitvale branch were not provided. Performance relative to Golden Gate is substandard, as shown below in Figure 4 for the current quarter: Metric | Manzana | Golden Gate | Average TAT | 6 days | 2 days | Renewals late | 44% | - | Renewal loss rate | 47% | 15% |
Figure 4. Performance of Manzana relative to Golden Gate
The absence of late renewals by Golden Gate is telling – that represents a yawning gap to Manzana's 44%. It's a clear indicator as to why Manzana loses close to half of the renewals segment of its business every year. These losses may also be attributed to the incentive of staff to work on new policies rather than renewals. The workflow is not optimised, with staff working in silos rather than pooling resources. These issues are reflected in Manzana’s profit and loss statement. Miscalculation of TAT
The way Manzana calculates the TAT has also led to major problems. Exhibit 3 provided by Manzana outlines the current processes to calculate the TAT. (See spreadsheet Syndicate 10_T2_2016_PT_Manzana Insurance Fruitvale Branch, Backlog and TAT sheet). Manzana assumes activities need to wait for earlier activities to finish. The assumption is that policies move from one activity to another in batches equal to the total number of current policies with that area.

As an example, the underwriting team only starts working after 0.6 days once distribution clerks have finished their work. This has resulted in a miscalculation of TAT. Manzana uses the 95th percentile standard completion time, which is too conservative. 95% is a much higher number than the average of 50%. These miscalculations have led to an incorrect TAT of 8.2 days. That is the worst case scenario. We recommend Manzana change how they calculate TAT. A lower TAT can attract more business.
Cost Structure and Prioritisation
Manzana’s profits have deteriorated over the last year. Commissions incentivise staff to prioritise new policies over renewals. Despite this, renewal policies have much higher margins than new policies. They also take less time to process. The high renewal loss rate of Manzana is driving profit down. This can be seen through: * Flat revenue from Q2 1989 to Q2 1991 * Increased gross revenue % from new policies * Increased commissions and expenses due to the higher commission rate on new policies * Increased operating expenses, where Plus Program accounts for ($97,000) or (49%) of the operating expense increase
For a full P/L analysis, please refer to the spreadsheet, Cost and Lead Time Analysis sheet.
The increase of new policies may have hidden affects on profit. One may be adverse selection: clients anticipating a loss may initiate new policies. Underwriters in turn may be too lenient in their assessment of that risk in order to earn their bonus. Manzana may also be forced to service a more risky client base because of their poor service level, they receive the customers that their competitors will not insure.
In 1991, the gross premiums for RUN was $6,724 and RERUN $6,205. Given that the agent fees of 7% commission will go to RERUN and 25% will go to RUN, prioritizing RUN isn’t ideal.
The gross margin for RERUN is higher at $5,771. The Gross Margin percentage is 93%. RUN will only receive gross margin of $5,043 with a Gross Margin percentage of 75%. From gross margin perspective, RERUN gives a better return.
The policy process time for RUN gives higher gross premiums, but the total time cost for RUN is almost double. RUN takes 258.6 minutes to process the request, where RERUN only takes 144.3 minutes to complete the request. This indicates the total time spent on process request for RERUN is much quicker than RUN.
Exhibit 6 provided by Manzana shows the average RUNs increased in the last two quarters of 1991 - 3.36% and 8.59%. Late RERUNs increased by 10.18%. Lost renewals increased by 13.34%. These are the key factors affecting the profit performance.
The incentive structure must reflect the desired behaviour of staff. It is recommended that Manzana flatten the commission structure to encourage a balance of renewals and new policies. It will increase margins and improve the service level. Fewer customers will leave the business because their policy renewal came late to their agent.
Utilisation
The utilisation analysis of the processing teams for distribution and underwriting can be seen below in figure 5 and 6. For a more detailed analysis, please see refer to the spreadsheet, in sheet Cost and Lead Time Analysis. There is a high utilisation in distribution (88.8%) and underwriting (70.4-96.9% across territories). These are bottlenecks. Given these departments are at the front-end of the process and feed the downstream process, the rating and policy writing department are sitting idle for 23.7% and 36% of the time respectively (please see figure 7&8). As such, the system performance is not being optimised. Given the reported backlog of orders, this would suggest that Manzana are not demand constrained, but that variability in the demand and process times impact the utilisation levels.

Policy Type | Distribution (4 clerks) | RUNS | 11.1% | RAPS | 41.6% | RAINS | 9.1% | RERUNS | 27.0% | TOTAL | 88.8% |
Figure 5, utilisation of clerks for distribution by policy type Policy Type | Territory 1 | Territory 2 | Territory 3 | RUNS | 13.1% | 8.1% | 7.1% | RAPS | 53.6% | 36.1% | 36.9% | RAINS | 8.2% | 5.2% | 5.4% | RERUNS | 22.0% | 29.1% | 21.0% | TOTAL | 96.9% | 78.5% | 70.4% |
Figure 6, utilisation of clerks for underwriting by policy type and territory
There is also an unbalanced utilisation across the territories. The division of territories across geographic lines eliminates the economy of scale benefits of pooling. Territory 1 has a high utilisation of 88.8% versus Territory 3 running at only 70.4%. That indicates that the spread of requests is not even between the three territories. There is 21.5% and 29.6% idle capacity in Territory 2 & 3 respectively. This variation in utilisation between the territories is resulting in higher renewal loss rates in territory 1 (39%) versus territory 2 & 3 (21% and 33%). Policy Type | Rating (8 clerks) | RUNS | 6.1% | RAPS | 26.9% | RAINS | 6.8% | RERUNS | 36.4% | TOTAL | 76.3% |
Figure 7, utilisation of clerks for rating by policy type Policy Type | Policy Writing (5 writers) | RUNS | 9.2% | RAPS converted to RUNS | 7.2% | RAINS | 9.0% | RERUNS | 38.6% | TOTAL | 64.0% |
Figure 8, utilisation of policy writers by policy type Lead Time The bottleneck in underwriting is caused by the priority given to RUNS, RAPS and RAINS. This makes a bad situation worse for underwriting RERUNS. The RERUNS are only released one day in advance. As they are not prioritised, it obliterates the chance of them arriving to agents on time. RERUNS are renewals of existing policies and are generated automatically so the lead time is completely controllable.
RUNS, RAPS and RAINS are time sensitive and cannot be delayed. RERUNS are not time sensitive as long as they are released with enough time to get them to the agent 30 days before renewal.
Other issues that contribute to the blowout in TAT for policies are the double counting during the RAP process and the overly conservative Standard Completion Time (SCT) at 95%. 95% of the sum of 5 random times is much less than the sum of 95% each time a policy is processed, even taking variation into account.

Recommendations
Manzana should seize this opportunity to improve service, reduce customer attrition and regain competitiveness.

1. Pool Resources in Underwriting
Territory 1 is operating at dangerously high utilisation levels, whilst the others have idle time. By pooling resources, Manzana will be able to realise scale efficiency. It will result in reduced wait times. It will also reduce department silos and encourage organisational learning.

2. Change Lead Time of Automatic RERUNS
Manzana currently generate RERUNS too late to arrive on time. Manzana can generate RERUNS 35 days before they're due for renewal and process them within 5 days. Processing within 5 days should be a KPI. It would ensure customers receive renewal advice 30 days before the policy expiry. It would also mean a higher rate of success and improve the perceived level of quality in the eyes of agents. Please fig. 9, and spreadsheet, Cost and Lead Time Analysis sheet for linked analysis. | SCT (99.99%) mins | Days | Agency Buffer (Days) | LT buffer (Days) | Recommended wait time (Days) | RUNS | 863.54 | 1.92 | 0 | -0.8 | 1 | RAPS | 344.21 | 0.76 | 0 | 0.0 | 1 | RAINS | 609.29 | 1.35 | 0 | 0.0 | 1 | RERUNS | 323.66 | 0.72 | 3 | 0.8 | 5 | Figure 9, adjusted Lead Time to manage variation in policy type arrival (wait time in minutes and days)

3. Advertise Average TAT of One Day
Manzana must compete with Golden Gate, and adopt a realistic way to calculate TAT. With a 99.9% confidence interval, RUNS processing time average is 1.7 days. Manzana can use idle resources from processing RERUNS, if they adhere to the 5 day lead time (see fig. 9).

4. Amend Commissions
The commission structure in place results in an agency problem. The behaviour of the underwriters in chasing RUNS to the detriment of RERUNS was not aligned with the goal of the company – to maximise returns. Manzana should amend the commission for Underwriters to a 16% flat rate, regardless of work type. This would mean there was no preference for RUNS.
Manzana should also consider incentives for the other departments, i.e. Distribution, Rating and Policy Writing. It would encourage all departments to share a common purpose. That would improve the collaboration and collective mind-set of the departments.

5. Introduce FIFO for RUNS, RAPS and RAINS
Poor planning and prioritisation hurts Manzana. Introducing FIFO for RUNS, RAPS and RAINS will reduce the backlog of policies. It will increase the flow rate of the process, as time isn't wasted reshuffling policies. It will also reduce the bloated wait time, as more complex policies aren't held back.

6. Reduce Variability in Processing Time and Customer Demand
Manzana know the schedule of RERUNS. They should track the volume each month, so peaks and troughs in demand can be better managed.
Manzana should also provide staff training to standardise processes. It will reduce the variation in skill levels for team members. That will result in a decrease in processing time variation.

7. Assign One of the Policy Writing Staff to Distribution
This action will improve the balance of idle capacity between departments. It would result in an increase in Policy Writing utilisation from 64% to 80%. It would decrease the Distribution utilisation from 89% to 71%.

Glossary of Terms
CV – Coefficient of Variation
FIFO – First in First Out
RAIN – Request for Additional Insurance
RAP – Request for Price
RERUN – Request for Renewal
RUN – Request for Underwriting
SCT – Standard Completion Time
TAT – Turnaround Time
STDV – Standard Deviation