Insurance companies

Identification and Anti-fraud Services

Within the insurance market there are a wide variety of issues that insurers face. These include an increase of fraud at all stages of the insurance lifecycle, cost rationalization, and access to impartial tools which insurers can use to set internal targets. Also the concept of sharing information to combat fraud and reduce insurers’ operational costs has gained more and more importance.

CRIF offers solutions for more accurate identification verification and anti-fraud management at the key stages of the process, including quotation, underwriting, claims handling and fraud investigation, to help insurance companies increase their profits, improving their knowledge of customers and efficiently managing risk assessment.

Insurance bureau services

The insurance bureau services developed by CRIF, allows the insurance industry to confidently assess the motor, home, personal injury, health, marine, travel and pet insurance history. Constantly enhanced to meet the demanding requirements of insurers moving fraud prevention and risk assessment processes upstream, insurance bureau services effectively supports insurance companies throughout the client lifecycle, from the policy quotation to the claim stage. Thanks to a technologically advanced platform able to handle the volume demands required for quote screening, they guarantees high performance with supporting service levels. Insurers have the flexibility to choose subsets of data services and information needed to build up specific risk and fraud prevention models, giving total control to the insurer.

The Insurance Bureau Service can also work in combination with credit and payment transaction data, as a strict correlation has been proven between bad payment behavior and propensity to claim.

Insurance companies may decide to contribute data relating to claims only or claims and payment transactions related to their customers. Data enquiry and ring fencing will be guaranteed by reciprocity rules.


  • Ability to increase premium revenue and profitability by charging rates linked to the true risk involved.
  • Improved quality of the business portfolio and reduced exposure by accurately identifying risks.
  • Reduction in the manual effort and costs involved in confirming at the underwriting stage whether an individual has consistently disclosed his claims history and can therefore benefit from a “no claims” discount.
  • Reduction in claims expenses and faster processing through more efficient claims handling.
  • Control of overall costs by identifying multiple and potentially fraudulent claims.


Sherlock is a fraud prevention product which allows insurance companies to manage the identification and fraud prevention stages faster and more efficiently.

Multilingual and multiline (supporting motor, home, personal injury, health, marine, travel and pet insurance lines), Sherlock can integrate any internal and external information source, and analyze the results using innovative machine learning and analytics tools. A unique and intuitive interface enables claims and policies to be quickly classified on the basis of the actual risk of fraud, checking personal identity, reporting anomalies, and performing customer intelligence activities.

Using tools which allow users to configure their own expert rules and multidimensional anomaly identification functionality, it is now possible to move from a verification approach based on the characteristics of the individual claim and with a limited number of variables, to the analysis of a combination of multiple variables which are difficult to analyze manually and using traditional algorithms. Therefore, Sherlock enables potentially unknown or not previously identified fraud scenarios to be discovered and individual cases to be examined in more depth, carrying out additional checks on related subjects, third parties and associated addresses through the ntwork analysis function, showing the identified links graphically. Through a simple and comprehensive report, it is possible to quickly identify any anomalies and areas of risk which require further investigation and to analyze the searches carried out thanks to the traceability of all the activities performed on each subject under investigation.

Finally, users can carry out interactive searches in real time, and view on a single screen the aggregate results of all the links generated from a wide rage of variables relating to individuals, companies, addresses, vehicles, e-mail addresses, telephone numbers, etc...which could also be relevant across different claims.


Elixir is used by major life insurance companies for screening, reporting and alerting users on the financial standing of brokers and independent financial advisers (IFAs). Elixir members supply information on debts, legal actions, repayment schedules and written-off amounts using a web-based application. The system helps insurance companies validate and monitor the entire lifecycle of their business relationships with individual IFAs, supporting confidence, transparency and efficiency in relation to commission credit and debt.

Relevant data is compliantly shared between Elixir insurer users to mitigate potential risks from unethical IFAs who may systematically target multiple insurers in order to commit commission fraud and IFAs who may repeatedly require extensive and long-term commission credit due to financial instability or propensity to be subject to legal action. This in turn protects insurers from reputation risk, and supports mutually beneficial relationships with their IFA partners.

Bid data Analytics and scoring

CRIF provides scoring models and expertise, empowering business analysts, from beginners to advanced modellers allowing them to develop, build, test, deploy and manage predictive models. 
Predictive analytics allow insurance companies to extract information from existing data in order to determine patterns and predict future outcomes and trends. It forecasts what might happen in the future with an acceptable level of reliability, and includes what-if scenarios and risk assessment.

Today, new methodologies, such as machine learning and Genetic Algorithms to solve problems and find indicators or Evolutionary Neural Networks and the availability of unstructured, social network, and behavioural data are used to optimise the relationship between information.

The application of knowledge discovery and data mining through the use of Link Analysis and Neural Networks enables identification, analysis and visualisation of patterns in data; embracing intelligent technology to auto learn and inform processes.

CRIF has been investing in research and development to design advanced techniques in risk profiling and develop our DNA of Analytics drawing on information about human behaviour, lifestyles and habits and computing large quantities of both structured and unstructured data available via social media platforms and beyond, dishonest behaviour, information discrepancy, to generate a more powerful and accurate predictive risk profile.