Best Practice

Internal Data Modelling for the Insurance Industry

Introduction

In an ever more connected world where information availability and integration are key enablers of business, industry standard models are still poorly understood and difficult to implement.

Insurers need to quickly understand their business and interpret data to foster innovation, reduce operational costs and provide a platform appropriate for accurate and fast assessment of risk / innovative pricing, for compliance and to leverage new sales channels.

Equally, there is a strong desire within many insurers to change from being relatively faceless policy-centric organisations into being customer centric organisations offering an ecosystem around their core proposition. Information and data are key enablers in changing insurers into organisations that are relevant in the digital age.

Glue Reply has extensive experience in the industry, helping clients achieve the best of both worlds – simple useable models and architectures that provide a common language within organisations, support compliance and customer centricity. Glue Reply also has extensive experience in helping organisations deliver industry best practice.

Glue Reply

Glue Reply is the Reply Group company specialising in IT architecture, integration and data solutions that drive value for organisations. Pragmatic in its approach, Glue Reply provides independent advice on the technology solutions that fulfil business objectives. Our core proposition is to help clients maximise value from their business change and technology investments by helping them define, design, implement and resource best practice.

Glue Reply helps drive operational excellence across the organisation, whilst preparing the client for digital transformation, cost reduction and data exploitation. Glue Reply delivers this consultancy to all sorts of organisations regardless of size, sector or complexity.

For more information please contact us at glue@replyltd.co.uk

Industry Standard models

Rationale against using Industry Standard models internally

  • Industry standards offer a superset of what is required (i.e. too much and too complex)

  • Industry standards usually do not address the full scope of an organisations’ proposition: usually Business differentiating elements of information are not adequately catered for

  • Version management and version upgrade is out of control and becomes difficult to manage; an internal structure insulates the business from this, requiring far fewer changes when the standard changes

  • Being in control of version-to-version mapping (not provided in standards)

  • Many of the industry standards are not ‘internally consistent’

  • In particular customer and parties are not sufficiently designed to support true customer-centricity

Rationale for using Industry Standard models internally

  • No reason to reinvent the wheel; thousands of hours have gone into modelling the core entities, definitions and relationships

  • All insurance companies have a need to both utilise information internally and exchange information within the industry

  • Following standards for information and data modelling enables further standards to be leveraged – such as the ACORD business glossary or the component model

  • New regulatory requirements will easily and consistently map to the industry standard models, providing a solid basis for implementing them

Target Engineering

Glue Reply helps insurance companies, to turn their logical data model into a physical data model. This is the essential step in the process to materialise the data model – turning theory into practice.

  • Creation of a standard methodology to take ‘subject-oriented’ slices of the model and turn them into a message structures

  • This process can support XSDs for services

  • This can also support other physical realisations such as CSV, Fixed delimited and JSON formats

  • Equally, it is possible to use the logical model for database design (physical database schema) through a standard forward-engineering technique. This might be used for additional/future master data repositories

  • The model can be used as part of the realisation of a data strategy, for use to support a Big Data initiative and can support a number of styles of analytical management

  • The model can also be used for dimensional modelling as might be required for reporting

Operating models and business events

Glue Reply are able to work with clients to map their business processes and business events to the model - enabling clients to identify understand the whole data journey throughout their information estate. Once the organisation is using a common model and common language, Glue Reply can help clients identify inefficiencies and opportunities for consolidation. Glue Reply also has a methodology to tie together business capabilities and business capability models into information models.

Realising the Information Model

Glue Reply has deep expertise in the Information Model development area and can realise models in a relatively short space of time. This contrasts sharply with experiences of many organisations who struggle to ever get to a high quality model or modelling approach or else require significant effort and time to build and maintain the model. Glue Reply can lean on experience in developing models for some of Europe’s largest insurers to save our clients time, money and de-risk the realisation. Glue Reply has significant intellectual property developed in this area in order to accelerate model realisation.

Glue Reply Approach - Best of both worlds

Glue Reply follow a top-down / bottom-up approach to information modelling. This leverages the standard definitions and core concepts from industry models whilst providing a bespoke design that is relevant to the modus operandi and unique customisations of the client.

When building a model, the top-down approach leverages:

  • Industry models that are relevant – this can be a wide range of standards, but of all the standards, ACORD is the one that is used the most

  • Insurance Message Models – there are a number of message models out there

  • Direct line of business standards – this can include specific standards that relate to a line of business, such as benefits using HR-XML, private medical using HL7 etc.

  • Other relevant standards – there are a few widely used standards such as OAGIS used for industry-agnostic lifecycle services. There are also a series of ISO- standards for information structures

  • Glue Reply modelling accelerators – proprietary well tested patterns that facilitate swift delivery

  • Simplicity – if the model is difficult to understand and control, it won’t be used. Glue Reply specialise in turning complex requirements into easy-to-digest models

When the model has been created, it is important to validate that it works and can be implemented within a realistic timeframe. The bottom-up approach draws on existing data sources such as systems, processes and data repositories and uses them to test the model. At this point, it may be necessary to provide localisations as extensions to the model where requirements, definitions and processes differ on a regional or structural basis.

The model is then mapped to the relevant standards in a way that simplifies integration in future and ensures that outbound messaging will be compatible with the likely external standards.

Last, the bottom up approach explores the maintenance effort of keeping the model synchronised and relevant to the business.

Picture

Glue Reply specialises in IT architecture, integration and data solutions that drive value for organisations. Glue Reply helps its clients succeed by turning strategy into tangible solutions and vision into practical outcomes. Glue Reply supports and drives clients’ objectives and enables organisational, process and technology change.