An investment manager is a complex web of data. In a data-intensive industry, few firms have mastered their reference data to create a single view of their “data truth” and even fewer firms have solved the single data platform to meet the needs of front, middle, and back office, distribution, and corporate functions. With data science being added to the mix, the data landscape has become increasingly more complex and also exciting, elevating data from operational efficiency to alpha generation. With data at the heart of all investment business functions, an under-performing data platform impacts business results.
The good news is that a well-designed data platform that is flexible, fast, governed and accurate is within reach. Advances in data technology, data mastering, data governance, and years of design experience solving investment management data challenges can be applied to lower inefficiency and gain increased value from your data. Many are put off from fixing their data platform and managing a large, complex and costly project. However, Olmstead advocates that improvements can be made incrementally if you have a good design blueprint and leverage smaller business initiatives wisely.
It may seem overstated that data is always a challenge, but the question many senior executives are asking is “what impact is our data platform having on our organization?” There are direct and indirect costs to organizational inefficiency, business results, and lost opportunities caused by an under-performing data platform. Olmstead refers to this as data tax. Inefficient and inaccurate data impedes employee performance impacting all roles from senior leaders, investors, compliance, risk, sales, quants, data science, accounting, and in short, everyone, who must accommodate it in their everyday work. This extra work required to gather, format and scrub data is time-consuming and expensive. Forrester reports that “nearly one third of analysts spend more than 40 percent of their time vetting and validating their analytics data before it can be used for strategic decision-making” and IBM estimates the yearly cost of poor quality data, in the US alone, in 2016 was $3.1T.
Data tax is the extra cost applied to all business functions that are dependent on data. A well-built data platform that provides answers, decisions, and desired outcomes has little to no increased costs and is therefore a minimal “tax”. Poor performing, disorganized, and inaccurate data platforms impact firms everyday across every role. Data tax can be broken down into four categories:
Data Inefficiency “Tax”
- Inefficient access to data and data wrangling to perform your job function
- Forcing business teams to do data integration work and data mastering
- Not knowing where the data is or what data is available
- Forcing analysts to create their own data marts in Excel
- Redundant data solutions for same analysis purpose across teams
- Buying the same data multiple times across the enterprise
- Overall data frustration in the organization
Data Project “Tax”
- Most data-intensive projects run significantly over budget
- Inefficient and disorganized data platforms cause project delays
- "Data Readiness" and data architecture get cut from the project budget becoming the primary factor leading to cost overruns and failed projects
- Project delays create more "data tax" as data solutions become tactical to meet 'data at all costs' (hacks causing data inefficiency)
Lost Opportunity “Tax”
- Did not understand or anticipate the client need or impact to new product development
- Lost revenue – missed communications, incorrect pricing of products
Business Risk “Tax”
- Inaccurate data, potential for liability, fines
- Multiple data results for the same question
- Insecure data platform, potential for hackers, misuse
- Incomplete view of a risk scenario, lacking data leading to incorrect decisions
- Reputation damage from poor data quality
For the sake of argument, take your entire payroll of your firm and multiply that by 10%, 20%, or 30%. Given that data impacts all roles, this is an estimate of your data tax. If we all could have built our investment business from day one with a modern data platform and a data architecture that considered all the data needs, then there would not be a problem. Instead we tactically added a trading platform, then changed the accounting platform, then added new asset classes, and soon enough we have compounded data inefficiency.