Data management:

mastering data makes the difference

The harsher operating conditions facing the investment management industry in the wake of the financial crisis have given rise to the need to reappraise existing ways of managing data. One solution is to provide consolidated, cleansed data in standardised formats based on client-defined rules.

By Martin Buchberger

Order your free subscription of the Journal here

Data (or information) man­agement systems have be­come an essential corner­stone for almost all the operations of an investment manage­ment organisation by providing quality, timely data for decision support. Systems have progressed to cover a wide variety of investment management areas includ­ing asset registration, financial manage­ment, process scheduling and control, materials management, maintenance management, condition monitoring, risk management, reliability management, and safety management.

Due to the myriad systems and unique combinations of information systems within organisations, past research into data management systems has targeted specific industries or specific systems. Research emphasis within investment management itself has been placed on a few select areas such as control systems, maintenance, condition monitoring, and reliability. This has led to a dispar­ity in the level of research into data management and information systems across the entire spectrum of the invest­ment management industry.

The last two decades have also seen an influx of the use of computer-based tech­nology into investment management as breakthroughs significantly increase their functionality and subsequent adop­tion. Advances in computational power have paved the way for harnessing com­plex algorithms for the analysis of opera­tion and condition data.

Research into database technology has allowed huge volumes of data to be col­lected and processed, as well as spurring on the advent of the data warehouse. The Internet has brought the benefits of in­formation-sharing and accessibility to the fore, and corporate system integra­tion and workflow management are now being addressed in current research.

Understanding the adoption and use of the aforementioned technologies allows the industry to formulate appropriate strategies on where to focus future re­search and development effort for in­vestment management systems. How­ever, with the availability of competing technologies compounded with a mix­ture of organisation technology adoption strategies, it can often be difficult to identify the current state of technologi­cal usage in investment management.

It is interesting to note that in recent years an entire industry has grown to address related quality issues found in corporate application systems. These technologies provide comprehensive and intelligent algorithms that pro­grammatically mend, consolidate, and attempt to repair the data resident in corporate databases.

Further, the realisation that data qual­ity is a major contributor to the overall cost structure of an organisation has led to major system initiatives intended to address this issue. System integration efforts among disparate application sys­tems, databases, and business processes are evaluated and when possible con­solidated to minimise poor data quality. All these efforts are critical to the ongo­ing operational effectiveness and corpo­rate agility of global investment man­agement companies.

QUESTIONS TO ADDRESS

To understand how data management in investment management can best be ap­plied in the corporate arena, investment management companies have to ask themselves the following key questions:

• What is the precise composition of data management and information systems in industry management op­erations?

• Why are some systems used while others are not, and what improve­ments can be made to current invest­ment management data systems?

• How is the success of an investment management data system measured?

• What is the level of integration be­tween these data systems?

• What data are regularly discarded and why?

• What is the level of investment man­agement data warehousing activities in company organisations?

Over the years, users of corporate infor­mation have come to recognise the age-old adage "rubbish in, rubbish out." For these reasons, programmatic steps are taken to ensure data quality. Data entry and field edits are imposed to catch the obvious typographical errors and data­base triggers and rules are instituted to prevent the entry of duplicate records. While these measures serve an impor­tant role in the information capturing activity, many errors, omissions, and duplicates can and do occur. In this context, data quality and data valida­tion are two separate and distinct steps necessary to ensure best practices.

If the financial systems applied by in­vestment management companies are examined, for example, they are in­tended to reflect the current state of a business enterprise. The transaction systems and accounting procedures are intended to interrelate such that an ex­ecutive can understand the financial condition of the enterprise at any mo­ment in time. The reality is that most systems disconnect between the infor­mation-gathering phase and the busi­ness process. Either the transactional data entry fails to support the level of detail needed to reflect the business process or the business process is not followed. The latter is more common.

Figure 1. Driving forces for reference data automation. Source: AIM Reference Data Survey 2010.

SOURCE SYSTEMS

As data management systems form the source systems to a data warehouse, it is important to understand their composi­tion within a corporate organisation. Data processing and reporting is inci­dentally the primary focus of data ware­housing by providing a platform for inte­grated data analysis. Most organisations have finance and account management systems, while the adoption of risk and reliability management systems lags be­hind. The ordering of the two leaders is slightly strange as most finance manage­ment systems also include an account management component or module. This is also true with investment man­agement, where it is typically a compo­nent within the IT infrastructure. One possibility could be that organisations are not purchasing these modules in their system packages, or that these modules are not being used to their full capacity and are hence discounted.

Data management systems are often built around a workflow specified by the sys­tem’s developer. Due to the lack of work­flow customisability in many data systems, an investment management company will often need to adopt the system workflow model, rather than adapting the system to the current organisational workflow. While this forced adaptation produces both beneficial and detrimental effects, in the case of investment management, the benefits seem to be greater.

What tends to be overlooked is the task of identifying best practices in reference data system selection, data warehous­ing, systems integration and data reten­tion. Key issues to be addressed here are: a significant adoption of informa­tion systems and data warehousing across different business lines; the pri­mary use of information systems to streamline business processes and en­hance reporting; and the strong desire for improved system integration for next-generation investment manage­ment information systems.

Past studies into data management within investment management tended to focus on understanding why specific systems or specific data management processes are implemented in a given company. There was little investigation into the broader picture of such invest­ment management systems, their com­paritative scope and their overall data integration strategies. Lacking was an exploratory, cross-sectional and inter­national survey that examined a variety of data management issues directly im­pacting investment management com­panies across the board.

AIM REFERENCE DATA SURVEY 2010

The AIM Reference Data and Risk Management Survey is designed to ad­dress some of these issues. Published every consecutive year since 2004, the seventh global survey published in No­vember 2010, and drawing on the re­sponses of 380 financial institutions from 51 countries, aims to provide in­sights into the driving forces, challenges and planned investments for reference data automation in financial institu­tions.

A special objective of the study, under­taken in the period from April to October 2010, is to take a close look at reference data management procedures and ob­serve the developments over the years in order to help institutions obtain a better picture of their business in a constantly changing environment. By comparing their own data management strategy with the regional or global results, enterprises are able to assess their future steps in ref­erence data management.

While the primary reasons for using investment management data systems are to improve business procedures and data reporting, Martin Buchberger, CEO at AIM Software, explains how data analysis and reporting can help companies in detecting process inefficiencies and provide a platform for continuous improvement.

The survey results for 2010 indicate that in the wake of the financial crisis, enter­prises consider the reduction of errors (76% of all responses) and costs (66%) as well as the management of risk (53%) as the main driving factors for reference data management (see Figure 1). Com­pared to the results of previous years, these figures show a steady increase of institutions’ awareness in these areas.

A major finding of the latest survey re­veals that managing corporate action is gaining increased importance as com­panies recognise the need for the effi­cient processing of corporate actions in a timely and reliable manner. More than one-third of all participating insti­tutions stated that they want to invest in the management of corporate actions in the near future, a number that has been growing steadily over the last few years.

In addition to the further automation of corporate actions, enterprises continue to focus on security master files (i.e. golden records or golden copies) to centrally man­age their reference data. Figures in this area demonstrate that the demand in this area is still on the rise. Whereas three years ago, only 38% of all respondents stated that they had a golden copy in place, the 2010 survey shows that 52% of all re­spondents currently feed reference data into a centrally managed repository. This confirms that companies are aware of their need to further enhance operational effi­ciency and to support growing risk man­agement and compliance requirements.

The survey results also indicate that in the new challenging business environ­ment, enterprises are continuing to take urgent measures to extend their reference data management solution. More than one-third of all respond­ents are currently working on im­provements in this area, whereas 16% state that they have already imple­mented a reference data management solution.

In the growing realisation among fi­nancial institutions that not enough is being done to prepare IT platforms for the demands of an increasingly com­petitive and regulated industry, a big push is seen this year and in the coming two years towards additional invest­ments in IT systems. A high 37% of all survey participants declare that they are currently working on extending their data management IT facilities.

WAYS TO PROCEED

In the post-financial crisis environ­ment, investment management compa­nies that want to succeed will have to consider several key ways to proceed. Among these are:

• focusing more on best-practice solu­tions that are capable of producing a fast return on investment (ROI, while still ensuring that solutions can be upgraded once the recovery gathers traction;

• coping with more regulations and in­creasing trading volumes with less human resources by applying best-practice solutions to help keep costs low with increased use of automation;

• responding to the increasing com­plexity of financial products and ad­ditional regulations will require more flexible solutions that can grow along with the needs of the company.

adopting modular and add-on solu­tions that can be easily deployed for standardising reference data systems and that subsequently can be ex­tended for universal application.

• understanding IT investments not only as a necessary spend item but also to reap benefits in both a quanti­tative and qualitative way i.e. by cut­ting the costs of data deliveries and improving the quality of the data that is being used by other core systems.

To ensure operational success and con­tinued expansion in this increasingly hostile climate, the main consideration will be to select and implement low-risk, best-practice solutions that can provide an immediate set of function­alities at the start of the implementa­tion. It seems that customers tend to choose providers with a strong track record to ensure that this provider will not simply disappear.

Data quality (or the lack of it) has been identified as a major contributor to enormous cost overruns and ineffective­ness. Lack of data quality is a signifi­cant deterrent for reducing operational profitability and effective financial management, leading to inaccurate de­cision-based information and other key operational inefficiencies. An entire in­dustry has developed in an attempt to deliver a systematic data scrubbing ca­pability to address this issue. However, data quality systems cannot always vali­date the data. This requires human in­tervention and a data validation proc­ess. Data quality in conjunction with data validation is the best practice for ensuring the maximum achievable ben­efits in investment management.

Through data validation, a company can confidently reduce operational inefficien­cies and costly errors. A wall-to-wall modular data management system achieves verifiable results while cleansing data. It provides the following advantages:

• it eliminates the duplication issue that often arises from partial and ag­gregate data recording;

• it ensures the quality of data in cor­porate investment management in­formation systems;

• It validates asset data required for compliance;

• it simplifies system deployments where information falls short of ac­curacy and validity;

• it provides a single point of entry for incoming data to be validated and processed.

CLEANSED DATA

Since the financial crisis, the investment management industry has become more risk-aware and its increased interest in en­suring cleansed, quality-assured data is a reflection of this. To gain traction in this sphere and deliver a more integrated and systematic approach to data validation, AIM Software has linked up with Sim­Corp in a global collaborative agreement. The agreement enables SimCorp to pro­vide AIM Software’s GAIN Data Man­agement software to its clients, in conjunc­tion with SimCorp Dimension, SimCorp’s investment management solution.

The data management software pro­vides consolidated, cleansed data in standardised formats, based on client-defined rules for use by downstream systems. As a result of the agreement, clients will be able to use the software to process securities prices, static and reference data and corporate actions no­tifications, with the resulting cleansed data being uploaded via a standardised interface. Consequently, clients can mit­igate risks of costly errors and avoid waste of resources associated with use of inaccurate data. They can also reduce cost and improve the accuracy of their data management processes through streamlined, automated workflows.

Introducing an integrated data manage­ment module in their IT systems allows investment management companies to streamline and overhaul their data processing capabilities. As a result, and directly due to the resulting improve­ment in data quality, companies can benefit from far more efficient processes and lower operational risk.

Further, they can reduce complexity of data vendor connections, as only one, mutual integration point for these ven­dors is required. Finally, companies have the flexibility to choose from among a wide range of data providers. In practice, this means that data can flow in from many different sources, collected in one single point of entry or repository, where it is scrubbed, cleansed and indexed be­fore it enters the user’s centrally man­aged security master data file.

INTEGRATED ARCHITECTURE

The majority of investment manage­ment companies use workflow manage­ment systems while life-cycle costing and risk and reliability management systems tend to lag behind in use. How­ever, these systems are becoming more a part of an investment management company’s information system architec­ture as adoption requirements increase for the business processes they support.

The primary justifications for using in­vestment management data systems are both to improve business procedures and data reporting. Streamlining business procedures through workflow automation decreases the overall time and resources required by each procedure, and thus, re­ducing costs. Data analysis and reporting also provides a method to detect ineffi­ciencies within processes, and provides a platform for continuous improvement.

While most investment management companies have already integrated some of their data systems through in-house development, easier integration is the most desired aspect for next-gener­ation systems. However, there is a lack of knowledge on data integration stand­ards within many investment manage­ment companies. As standards-based integration decreases the risk of adop­tion in long-term usage scenarios (at the expense of an increased initial cost), further awareness about integration standards needs to be disseminated.

The need for investment management data warehousing now appears to be firmly established in companies’ busi­ness considerations. Naturally, adop­tion has increased over the past decade and the primary reason for adoption is, again, the need for enhanced data anal­ysis and reporting. With the significant uptake of data warehousing for invest­ment management, it appears that this approach is a step in the right direction, although issues still remain on how to integrate data across different invest­ment management areas.

Overall, the main conclusion to be drawn is that the use of data manage­ment software for investment manage­ment in general is yielding favourable results for most users. Technology is be­ing used to automate processes leading to greater efficiencies, and complex data analysis is now becoming mainstream after decades of simple data capture and reporting. There is no clear-cut sector or size of organisation leading the charge; nevertheless, all investment manage­ment companies are residually benefit­ing from the ones that are.

Martin Buchberger has been chief executive of AIM Software since 1999. With over 10 years of experience in the data management sphere, Martin has worked as a senior project man­ager and risk management executive, includ­ing stints at Reuters Vienna before becoming CEO of AIM Software. He holds a Master’s degree from the University of Economics and Business Administration Vienna where he majored in Financial Markets, Operations Research and Information Technology.

AIM Software is one of the lead­ing providers of data manage­ment solutions for financial mar­kets, with offices in Switzerland, Austria, Luxembourg, France, the USA, Hong Kong and Japan. Established in 1999, AIM Soft­ware provides internationally es­tablished software solutions and services with more than 120 ref­erences in 16 countries. Sup­ported by its large client base, AIM Software offers low risk and low cost all-in-one software packages, based on its industry proven data management soft­ware platform GAIN. More in­formation at www.aimsoftware.com.