Surveillance & Monitoring Behavioural Analytics

4 min read

For many years, trade surveillance has been perceived as a secondary control process using technology that has the ability to identify inappropriate trading activity based on market abuse principles

Behavioural Analytics: The Evolution of Trade Surveillance


This approach has led to the creation of numerical parameters that are used to automatically generate daily alerts by various surveillance systems and investigated by supervisors or compliance officers, which can be slow and labour intensive.


If a genuine control breach exists, or a “near miss” is discovered, then these events are addressed according to internal policies and where necessary, reported to regulators.


Increasingly stringent regulations


In recent years, regulators have become more demanding on how banks and financial institutions monitor their trading activity. Regulators now have higher expectations of conduct and culture which require firms to monitor behaviour along with transactional control points related to the trade-lifecycle.


These behavioural expectations are clearly evident in the regulation replacing the old Approved Persons Regime, namely the Senior Managers & Certification Regime and further reinforced by new fitness & propriety guidance and conduct rules. If this isn’t enough evidence of the significant step-up expected in surveillance & monitoring capabilities, then consider the explicit market abuse prevention requirements embedded in recent regulations such as MiFID II, MIFIR and MAR.


Regulators now have the expectation that financial institutions need a more proactive approach to Trade Surveillance. This approach focuses broadly on a range of data points that influence the trade lifecycle from its point of execution to settlement and reconciliation. Financial institutions will need to use behavioural analytics, coupled with alert monitoring and data pattern analysis to anticipate potential market abuse and inappropriate trading events. This is a more preventative and forensic approach where the utilisation of data from diverse sources plays a huge part.


Behavioural analytic systems within financial institutions


Fortunately, much of the data needed to embark on behavioural analytics already exists in most financial institutions as a result of investigations on rate setting scandals such as LIBOR and the above mentioned new regulations. A behavioural analytical system will be able to analyse all control points of a transaction, use analytics to identify patterns of data, as a minimum covering:

  • Control data - e.g. cancel/amend reports, settlement reports, dummy and wash trade reports etc.

  • Transactional data - e.g. trade volumes analysis that can detect abnormal trading activities.

  • Reference data - e.g. counterparty data to link the information (using link analysis).

  • HR data - e.g. mandatory block leave, system access logs, physical access activity and proximity analysis.

  • Control room and communication surveillance data.


Challenges to be overcome in adopting a behavioural analytical approach


Adopting a broad behavioural analytical approach to monitoring & surveillance is, arguably, a quantum change in scale and complexity for any firm to undertake. There will be many challenges to overcome during the journey. Some of the key challenges are outlined below;

  • Relevant use cases - Designing and applying adequate and relevant use cases. There is no point building analytic surveillance tools if the right scenarios are not designed or appropriate for the business.

  • Appropriate skill set - Having the right business and technology skill set to understand the data requirements, select the right behavioural analytic tools and work with external partners to integrate and ingest data.

  • Appropriate exception reporting - Appropriate exception reporting and customisation for generating reports that provide transparency for future control process improvements.

  • Data privacy approvals - For personal and customer data it’s important to make sure firms have the appropriate legal approvals, access and retrieval controls (particularly given more stringent regulation on data privacy), potential cross border data issues and potential external (cloud based) solutions.

  • Data quality and data governance - The use of data quality teams to carry out data completeness and quality of data when utilising data from multiple and diverse sources.

  • Funding - A comprehensive behavioural modelling system requires an enormous amount of effort and expertise, using huge volumes and complex data etc. The cost of the leading edge software and the long-term infrastructure changes alone will require significant multi-year investment.


M3S Consulting are well positioned with experienced and committed practitioners to support you in achieving your ambitions for your surveillance & monitoring functions. Contact us, we are happy to discuss any aspects of Behavioural Analytics or Surveillance & Monitoring.

Knowledge Base