As promised, I have more on the subject of data onboarding. Last week, I listed a number of questions to ask yourself, the first (and arguably the most critical) of which is “What data is it?” In other words, what should the data comprise specifically?
The data we see at GoldenSource is everything from classic security reference data of all asset classes to data across all domains, including entity, legal hierarchy, ratings, classifications, ESG metrics, sustainability factors, cross-referencing, calendar, socio-economic, corporate actions and pricing.
But typically, for a functioning security master program, the key identifier schemes and classifications are the first datasets that need to be established, along with all the standards that need to be leveraged (e.g. ISO standards for currencies, countries, etc).
There is always at least one common or proprietary ID scheme which the front or back office is relying on heavily. Typically, there is more than one out of ISIN/CUSIP/CINS, the SEDOL with its OPOL (Official Place of Listing), the RIC, FIGI or BBGLOBAL/BBUNIQUE that is essential for quoting, trading, reporting, or just to speak with other internal systems.
And then there are security classifications, most prominently the CFI code (ISO 10962), the ISDA code, and the CIC (Complementary Identification Code). These are crucial because they may be the one canonical criterion upon which to base processing decisions and reporting. Virtually every data source has their own proprietary classification, too. Relying on those or a historically evolved, in-house classification alone can become problematic down the road, as they often reveal some shortcomings when used in a certain context.
The goal is to validate the data using these key identifiers and classifications right away, and avoid building a foundation of questionable quality. Getting these basics consistent early on is paramount because those are particularly difficult to clean up later on.
Ratings, prices and even issuer information, while obviously important, are really. secondary to these key identifiers and classifications, and can be more easily augmented later, provided these instrument essentials have been well defined and well maintained first.
Next week, I’ll have a ‘part two’ answer to “What data is it?”, when I talk about entity master data and its role in entity matching. Until then, don’t hesitate to email me with any questions you may have about your data onboarding work.