One of the previous JDX for Insurance Thinking Space articles, A Target Operating Model for the London Insurance Market, describes some of the challenges currently facing the London insurance market (the “Market”) and some of the solutions that are being developed and implemented in order to address those challenges. Of the initiatives included in the London Market Group’s Target Operating Model (TOM), a number have been highlighted as priorities in 2016. The one that will be the focus of this week’s article is Data Integration, more specifically Structured Data Capture[1].

Current state of slip data held in the Market

Over 175,000 contracts are written in the Lloyd’s market every year[2]. Roughly 250 brokers bring this business into the Market which they do using a document called a Market Reform Contract -traditionally known as a slip. A slip describes the risks that the carriers assume (in exchange for premium from the (re)insured) and these contracts come in a number of shapes and sizes. In theory, these have been produced in a certain way since the introduction of a set of standards in 2008. Once a contract is completed, some of the data from that slip is manually extracted from the document and keyed into systems by each of the parties to that contract.

The Market is a subscription market and therefore multiple carriers can underwrite a given risk. However, each of the carriers underwriting those risks will have a different risk appetite and different risk systems where data is recorded upon completion of a slip. As a consequence, once the slip is finalised, what often happens is that the carriers take identical documents away and manually input the same information into their separate in-house systems. All of a sudden you find that, whilst there are 175,000 plus contracts written every year, data from these slips is probably entered into systems up to a million times (perhaps more even). The contracts themselves are then scanned and stored in .pdf format.

Main challenges to capturing the data held in slips:

  1. Format of the Documents

In practice, the brokers (who put the contracts together) each have their own way of doing things. The ‘Market Reform Contract Implementation Guide’[3], is the standard for creation of slips and it contains a template document. Unfortunately, many have used this template to set up their own personalised template document. The downside of this is that the slip template was created using tabs (as would have been done in the old days on a typewriter) giving the headings on the left and the clauses on the right:

Some brokers have created their own template documents, which appear to have a similar structure, but which is actually a two-columned table. Therefore, any technology used to capture the data held in the slips needs to be able to have extraction rules based on tables and extraction rules set up to look at data separated by tabs (as in Figure 1).

  1. Different Users of the Information

As well as the broker who creates the slip, there can be multiple carriers (essentially parties to the insurance contract) which might want access to the information held within the slip. This poses a problem when trying to extract data from the contracts as each party might have their own preferences about which information it is that they want to take out. This might be because parties want to capture different fields entirely or it might be that one particular field is captured in a slightly nuanced way by different users.

The Benefits of Slip Digitisation

There are a number of benefits to be gained from the digitisation of slips. Most notably this digitisation would reduce the duplicated effort of manually re-keying data into various downstream systems by each of the parties to the contract. As a result, this could reduce transaction costs and ultimately contribute towards making the Market more competitive globally. Automation of the majority of this process would also have the potential benefit of increasing the accuracy of the data being captured (any automatic data capture would need to be quality assured of course). Greater accuracy of data would only serve to increase the accuracy of underwriting and modelling.

In addition, having better visibility of all active contracts allows the parties to have a better understanding of their current positions thus providing them with an opportunity to improve their risk management. Improved analysis of the contents of the slips can be undertaken allowing parties to be better informed when making strategic decisions.

Finally, through development of technology to extract information from slips, another possible upside would be to unlock ‘hidden data’. This hidden data is currently held in the slips but is only accessible via manual searches of what can be fairly lengthy documents (they vary in length but can be 40-60 pages long). By turning the content of the documents into fully searchable data, all of the information within a particular document can be found at the touch of a button. It therefore follows that slip digitisation could also help expedite claims processing.