Delegated Underwriting Authority roundtable: Our 10 key takeaways

1st March 2024

Davies hosted a Delegated Underwriting Authority (DUA) roundtable, bringing together leaders from across DUA value chain (Lloyd’s and non-Lloyd’s) and the Lloyd’s Market Association (LMA). The panel of experts reflected on the challenges and offered suggestions to improve the current DUA processes, data, regulation, and technology.   

10 key takeaways from the conversation are captured here.  

1) Larger, more established insurers have bigger challenges

Newer firms can invest in technology from the outset, giving them the advantage of newer, more agile systems and processes designed specifically for the needs of DUA processes.

However, more established insurers are likely to have a DUA operating model which has naturally evolved slowly over time, often as an add-on to more traditional ‘normal’ underwriting business. These firms are unlikely to have already invested in a complete overhaul of infrastructure, instead configuring DUA processes and systems ‘off the side of the desk’ as their needs develop. This doesn’t preclude substantial improvements in processes and data being achieved with this legacy technology, it just needs a different and considered approach to extract value.

2) Manual and inconsistent processes are inhibitors to process improvements

Despite a move towards greater automation within the insurance industry, processes, information flows and communication within the DUA value chain are still very manual. This can not only cause delays, but can also be the root of simple, yet serious, errors through copying and rekeying information from Word, PDF and Excel documents into numerous internal systems used during the DUA process. Intelligent automation can triage and streamline underwriting processes, allowing the processing of vast amounts data in real time, while, at the same time, significantly improving service delivery.

3) Reporting and insights are key data challenges, but it starts with poor underlying data

The state of underlying data is perceived to hamper AI adoption. Whereas in fact intelligent automation can greatly improve underlying data quality through joining up data sets contained in separate systems, reducing manual rekeying, and even augmenting the base data with new data points. Successful firms will be those that have real time, actionable insight, with a true understanding of the customer and their needs, and the ability to look forward and predict, not simply understand, historical performance. 

4) The bordereau is an integral part of the DUA process but is often not fit for purpose

Bordereau reporting remains a key component of the DUA process but despite this, quality is generally poor, particularly on premium and risk bordereau. 

More encouragement and support needs to be provided to MGA/Coverholders to improve the quality and timeliness of these bordereau before submission through to insurers. Ultimately, real-time data exchange is the aspiration for all, which would also reduce the lag between an insurer covering a risk before they know about it.  

5) A skills gaps is emerging

The insurance industry is increasingly turning to technology to optimise processes, however the increasingly rapid uptake in technology should be taken on carefully. Many are concerned on how replacing technology for manual processes could deskill workforces which could have a long-term impact on attracting and retaining talent into the industry. Firms have a responsibility to equip employees with the skills required and this will have a long-term benefit.

6) Available technology is holding DUA back

There are deficiencies with current ‘off-the-shelf’ DUA systems. So far, there is no one-size-fits-all, which is further compounded by insurers having to consider different regulatory regimes and legacy systems entangled in a multitude of other systems.  

Effectively, this means an off-the-shelf system is often not the best approach, with a more considered, and often less expensive, modular approach, preferable. Effectively this means taking the best components and integrating them together as opposed to developing an entire new system.  

While historically AI has been considered to improve operational efficiency, Generative AI can also provide significant customer service benefits, especially for those firms handcuffed with legacy systems. 

7) MGA/Coverholder audits are ineffectual

An audit of an MGA/Coverholder is generally undertaken as a tick-box exercise. The origins of this problem are rooted in the fact it is only an annual exercise further compounded by poor underlying data. Files are selected to manually audit, often with no strategy behind the selection. Additionally, this audit only covers a very small percentage of the cases in the binder/scheme. The ability to understand the full picture of the business and its processes through this one off, random sampling, is limited.  

The audit process needs to be reconsidered to be continuous across the year and to look at all transactions and use insights gathered from this to identify files for manual review. AI is the tool to help with this. Currently, there is a negative stigma associated with these audits. However, MGA/Coverholders need to be encouraged to use the insights gathered to improve business processes and deliver better outcomes for end customers. This would, in turn, create a true partnership between insurer and MGA/Coverholder. 

8) Contracts are at the heart of many problems

Continuous contracts are at the heart of many problems within the DUA chain as information within them is quickly outdated. This is compounded through the information associated with a contract being stored across the organisation by numerous people in Excel sheets, MS word documents and even desktops, meaning a single source of up to date, contractual, information is hard to find. A central document repository is a clear solution. 

While digital underwriting solutions can help solve these issues, they come with their own challenges. One of which is how insurers will ensure that contracts are kept up to date when automated solutions are relied upon. The launch by LMA/LIIBA of computable contracts in late 2023 is a step towards solving this problem. 

9) Complaints are not being recorded properly

Despite complaints being an area of regulatory focus a, complaints aren’t always recorded correctly, particularly in the company market.  

A robust complaints process, supported by effective mechanisms to capture solicited and unsolicited feedback, and multiple channels to confirm customer understanding, are essential to demonstrate good customer outcomes on DUA journeys and allow accurate, evidenced, regulatory reporting. 

10) Conduct risk within the value chain is being overlooked

Conduct risk is one of the single greatest day-to-day operational risks a firm faces, yet is one of the least understood and prioritised area of focus. A firm must consider the conduct implications of all activities within its extended value chain and be able to prove that it recognises both behavioural risks and poor practices—as well as putting in place procedures to mitigate these. This allows firms to show that they are delivering good customer outcomes in accordance with the FCA’s Consumer Duty requirements.

At Davies, we support the management of the delegated underwriting authority lifecycle to help you meet your business needs. Want to find out more? Contact Neil Strickland today to discuss how we can support your business with our Delegated Underwriting Authority solutions. 

Neil Strickland

Business Development Director – Insurance

Email | LinkedIn

    Keep up to date with Davies