Would you like a playbook setting out exactly how you can consolidate investment data in real time and how to get centralised access to your investment and adjacent data using data virtualisation? Do you want to understand what data virtualisation even means and how it can benefit your business?
In this blog post, we will walk through a number steps you can take to get immediate value from a consolidated source of investment data and centralised access to adjacent data.
Before we drill down into the details, let’s take a step back and look at the backdrop which frames the discussion. Active investment managers are facing challenges on multiple fronts from remote working during a global pandemic to increasing fee pressure and regulation around managing the investment process. The investment management system landscape is also facing a major reckoning from the growth of complex asset classes to tracking dozens of data points around ESG and sustainable investing creating the need for specialised solutions.
With the growing complexity of bespoke solutions it’s helpful to pause and ensure all trading, execution, reference-data, research and any other contributing systems combine into:
- a consolidated book of record where there is a single source of truth internally;
- a data set that has all of the attributes relevant to the investment process;
- a data set that includes all of the market and reference information used in the investment process (e.g. prices, benchmarks, ratings, etc.);
- and most importantly an accurate and consolidated source of risk and performance.
We call this a consolidated book of record.
It’s the dream state but difficult to achieve. Let’s look at counterparty data as an example – a specific area fraught with multiple access points, data types, identifiers, and access methods. Multiple counterparties and vendors provide data that you need for your core investment systems for valuation, position information, and cash availability and projections. Most asset managers use multiple vendors to provide different types of data such as transactions, balances, and prices. This brings with it a couple of problems:
- Counterparty data comes in different formats, different times, and with different identifiers.
- Consolidating all of this data into a single book of record, and preserving all of the original formats, identifiers, and data fields is critical when firms start using this data for things like reconciliation.
The operational playbook – consolidated investment data
Investment managers need a consolidated book of record that enables all of the counterparty and market data to be centrally stored, reconciled, and expanded as required to gain investment insight. What features does this investment data store need to have?
- Time stamp: Your investment data store needs the concept of entry date - a timestamp when the data is entered into the system. This goes beyond the basic business context of a price for example. You will have a price that is for a specific investment and date in a business context, but another date that is specific to when the investment system knew about it.
- Zero data mapping: From a technical perspective, the ability to centrally store all investment data should require minimal mapping and conversion, e.g. BY, buy, 112, by, and BY all mean the same thing and can be specified once. This moves beyond the simpler Export, Transform, Load process (ETL) because it prevents forcing the source data into a specific format that can cause loss of other data. This happens often when translating custodial data for example.
- Interoperable: Securities can be stored with every identifier used by counterparties and data providers (e.g. ticker, SEDOL, CUSIP, ISIN) and be referenced by any unique combination of identifiers. This enables you to uniquely identify an investment because you are not potentially missing a specific identifier used by counterparties or market data providers.
- Single source of truth: You can consolidate positions and cash so your front and back-office have real-time access to the same information for decision support, analysis, and trading. They have the confidence that the information is correct. If you store data with the concept of entry date in an immutable data store, this lets you reproduce data and calculations ad infinitum because it is never deleted.
Solving centralised access
Centralised access to investment and adjacent data lets you build a holistic picture from data that is classed as non-core to your investment platform. For example, centralising access to ESG data, alternative data, or CRM data.
We typically see firms use three ways to solve the “Centralised Access Issue”:
- You can copy your data into a centralised data store (e.g. data warehouse or data lake). This is a common IT approach but comes with some significant costs and problems. (e.g. the need to constantly feed in your data and keep it up to date, recreate entitlements, etc.)
- You can copy portions of your data when you need it into temporary stores. This is generally part of a manual Excel heavy consolidation process.
- You avoid copying the data and instead pull the subsets you need dynamically in a distributed and purely transient fashion.
But what if you didn’t put all of your data into one system in the first place? We see firms put data into a centralised repository because that source feeds reports, serves as an audit trail, feeds business intelligence, CRM, and is a catch all for sharing data to systems that are not necessarily time sensitive. The future of solving for these processes is to give access to the data you need, in real-time via a single entry point. This is called data virtualisation
What benefits does data virtualisation provide over a data warehouse or data lake?
- Query from single source without data duplication: You are able to efficiently query data from a single source without the expense of duplicating and normalising data multiple times.
- Simultaneous access to data warehouses or data lakes: Multiple operational data stores can be accessed simultaneously and have results combined with a single query. Again, this occurs without creating yet another copy of the source data.
- Centralise access and entitlement: Centralised access and entitlement prevents the costs and challenges around centralising access to data warehouses where those rights need to be duplicated from source systems along with the data.
Consolidating your investment data in a single book of record ensures data is never lost and can always be reproduced when it is stored with an entry stamp into an immutable data store. This consolidated data store, if done correctly, lets your upstream and downstream systems feed and consume investment data in real-time. Imagine the time you can save by removing the need for manual manipulation and delayed batch processes. As a result, you can increase efficiency by giving your entire firm access to the same view of your investments: You can now view the same investment book as your counterparties and can perform functions such as advanced reconciliations, analytics, and what-if analysis.
Taking a centralised access approach to your investment data lets you centralise risk, performance, security, CRM, alternative, and investment adjacent data. Combine this with a virtualised data engine to enforce entitlements across granular data, roles, and individuals and you have laid the operational foundations to prepare your fund for the challenges of tomorrow.