Amlin: Reducing the Risk & Cost of Using Spreadsheets

Amlin relied heavily on spreadsheets as a data-gathering tool, with the goal of achieving rapid response to constantly updated insurance scenario modeling information. However, the use of spreadsheets to input data from multiple users resulted in a workflow which suffered from frequent errors, and propagation of those errors, as well as the excessive elapsed time in the aggregation of multiple spreadsheets from the end-users. Furthermore, as the spreadsheets often contained highly proprietary information, they posed an IP exposure risk as well.

Using EASA, Amlin was able to significantly reduce the introduction of errors while greatly enhancing the efficiency of usage and protecting critical intellectual property.

As in any organization, many key processes at Amlin are under-pinned by spreadsheets, largely because spreadsheets are easy to create, familiar to users, and flexible. However, lack of control means that the risk of mistakes and the cost of executing the process are unacceptably high.

Staff at Amlin identified one process particularly in need of better control – RDS, or Realistic Disaster Scenario.

RDS requires every agent to submit a six-monthly report to Lloyd’s, showing exposure against scenario and by insurance risk code. In AMLIN’s case, this means some 20 underwriters around the world must each submit reports.

Historically, this process has been spreadsheet-based; AMLIN underwriters complete and submit a spreadsheet cloned from a template.

Complicating this process is the fact that the data required by Lloyd’s changes on a regular basis. Hence a new template must be created, tested and distributed each time there is a change.

The issues which arise include:

• Users change the spreadsheet to suit local requirements before completing and submitting, which results in aggregation errors down-stream.
• The spreadsheet must serve every user, and is therefore more complex than any one user needs – and most users have access to much more of the spreadsheet than is necessary.
• Any change requires that the new spreadsheet is distributed to every user – with instructions to remove the out-dated version. There is no mechanism to ensure the correct version is being used.
• Aggregation is carried out using a series of linked spreadsheets. Links have to be constantly updated, and many layers of spreadsheet refreshed.
• The data exists in many spreadsheets and not in a database, so historical comparison is time-consuming and expensive.

Reducing the risk and cost

A solution requiring the complete elimination of Excel was considered. However, that would have required significant investment in building a database application, re-creating the business intelligence already contained in the spreadsheet.

Instead, EASA’s spreadsheet management solution offered a far more cost-effective alternative – allowing AMLIN to secure a master version of the spreadsheet on a server.

Users now access it only via a custom web application created with EASA’s lo code application builder, allowing a more natural work-flow.

• The custom web application is so intuitive that training is no longer required.
• Users only see what they need to see; they no longer access the spreadsheet directly and are not able to “dabble” with it.
• If a change is required, it is made in one place and is immediately published to all users; version control is ensured.
• Aggregation is now automatic.

The immediate effect is to dramatically reduce the learning curve and facilitate deployment of key legacy software tools.

By avoiding the need to completely re-develop and then maintain new custom tools using conventional programming technology, Bayer estimates that EASA has saved the company several million dollars.