This article was written by Robert Simek, Head of Product for Sell-Side Analytics at Bloomberg.
The digitalization of capital markets is generating a second-by-second tsunami of data that financial institutions are using to better inform their trading decisions.
Across asset classes, trading information is being captured and then fed back into front-office workflows after identifying new opportunities and new risks that can shape the next transaction.
But the benefits are not being felt equally across the financial ecosystem. Traders in equities, derivatives and other instruments that are transacted across transparent exchanges are receiving crucial information on their activities every second. Those dealing in fixed income and other securities traded over the counter, however, must rely on their own efforts to capture vital intelligence.
The challenges are even greater for sell-side market makers, who still rely to a large extent on manual workflows. They work on their own pricing; the composite data that can empower buy-side traders is of little use to their sell-side counterparts, who need to understand a welter of proprietary information – such as current risk, liquidity from their client base, where trades took place, at what price and amid what bid-offer spread.
Capturing that data can be difficult and costly; the likelihood is that a firm will need to engage multiple vendors, and with no common data layer their feeds must be processed to fit into existing proprietary systems. And as there aren’t many vendors that are able to provide all the necessary capabilities to do that, these processes will need to be carried out in-house or by yet another third-party provider.
Without that data, however, sell-side organizations can find themselves left behind. They’ll be unable to automate and create rules that will help inform future trades. Unless firms have the data, properly processed and easily accessible, they will find themselves cut off from the sort of activities and the advantages that their competitors have.
Transformation challenges
The data journey can be broken down into several critical stages, all of which will lead to the creation of data sets that approximate the sort of time-series intel that is so critical to other parts of a trading enterprise.
The first step is to obtain the raw trade data. Optimally, that would include data on non-completed trades. This information can be just as potent a source of insight into client sentiment and intentions as completed trades, but is often overlooked or discarded. Few companies have put in place processes to capture this data, often because it has been scribbled onto a notebook or simply never recorded.
There are several ways in which such data can be snared. Electronic trading logs are one source. These become problematic, however, when a firm trades across multiple venues, all of which might use different terms to describe non-completed trades. This may lead to duplication of transactions in the data record, potentially contaminating transaction records.
Even more complicated is tracking non-completed trades that originated manually. That data resides in digital communications between traders and their counterparties, from which important nuggets of information must be mined using sophisticated natural language processing (NLP) technology and software. Having the capabilities to identify and retrieve that information is proving to be the final piece in the digitalization puzzle for many firms.
Once the data has been onboarded, it must be cleaned – normalized and made useable by the analytical processes that will reveal trading patterns and insights that can turbocharge the next deal.
Part of that process involves eliminating extraneous or duplicated information that can confuse or skew datasets, as mentioned earlier. It also involves mapping data to associated trades or instruments to create as complete a picture as possible of each trade.
Transactions don’t happen within a vacuum; the circumstances and outcomes of a transaction can be influenced by any number of external factors, including market conditions, price volatility, client sentiment and geopolitical situations. This ambient, or contextual information, needs also to be built into the digital record of each trade.
One valuable way of doing this is to link the financial instrument to an associated equity ticker, where greater data sets are made available, including social media sentiment and other “alternative” sources. Another method is to capture point-in-time data related to the exact prices and conditions of the market at time of trade, which can be utilized later for analysis.
Read The four-stage journey to Sell-Side fixed-income automation by Robert Simek for more insight on the automation journey.
Trusted partner
Onboarding that data and making it useable is only part of the challenge –ensuring it’s properly updated so that it can continue to be of value is a whole other challenge that all but the biggest financial institutions will struggle to overcome in-house.
That’s why entrusting that work to a third-party technology and data provider like Bloomberg will be valuable.
Bloomberg Sell-Side enterprise solutions can capture companies’ data, normalize it, enrich it with contextual information and map it correctly so that clients end up with a clear and ordered view of their trading activities. Importantly, that includes data on non-completed trades.
Once cleaned and structured, firms can then apply powerful compute and analytics to their non-completed trade data which is secured in the cloud and mine critical insights – such as performance metrics, liquidity matching and even AI-driven pricing tools – that will better inform investment decisions and drive revenue growth. And with built-in support for all EMS and trading protocols, clients can plug in from any on-premises setup.
To get the most value from their data, it’s important that sell-side firms don’t underestimate the difficulty they might face in their digital transformation. Having a partner with a long history of providing solutions to those complex challenges will make the difference between a successful transition and a costly, damaging one.