Big Data, Information Services and the Power of Knowledge

Since the term was first coined, ‘Big Data’ has become an industry buzz word, whose original meaning and potency has faded somewhat with overuse. Investors and entrepreneurs alike have long been chasing big data plays with little thought as to what such a term actually means. Simply put, ‘big data’ refers to the terabytes of data that technology has enabled us to produce, collect, store and mine. But in the majority of instances, data in itself has limited value.

The majority of data sources are commoditised. Numerous platforms have emerged in turn to crunch through these, in search of some predictive insight. Data-mining, cleansing and visualisation are all relatively simple tasks for talented start-ups as well as the behemoths like IBM, though most continue searching for the big data holy grail.

In our view, data is an enabler, rather than the opportunity itself; a means to an end. That valuable end is knowledge, such that if acted upon, it enhances revenues, lowers cost, or releases capital. Creating strong ROI and stickiness, such knowledge becomes a must-have instead of nice-to-have.

Over the years we've seen many successful entrepreneurs deploy different business models to achieve these highly desirable outcomes. Much is said about ‘unicorns’ in the VC world, but less is said about the fact that probably the largest set of multi-billion dollar companies originates in the Information Services sector: Acxiom, Bloomberg, Cap IQ, Experian, IHS, Markit, Thomson Reuters, Verisk, and the list goes on.

Historically, many of these giant leaders employed large labour forces to manually capture and aggregate disparate data sets. Usefulness and value resided largely in the breadth of the data set. In today’s world, new creative methods have allowed automated capture and processing of information; one can point to the examples of business models such as Factual (meshing and augmenting data sets), Waze (crowdsourced / collecting data by usage), and AI/machine learning (SwiftKey).

Trusted intermediary.

Another variety that is of particular interest to us is contributory models, where participants are self-incentivized as customers-to-be to contribute their proprietary data sets. Some refer to them as ‘give-to-get’. Labelled so because participants are compelled to give their own data asset to a central, trusted intermediary because of the perceived value of what they’ll get in return. The intermediary’s role is to clean, anonymise and give structure to the contributed data and to provide back a market or consensus view that removes opacity from the market. What the participants pay for is the transparency, accuracy and timeliness the intermediary has created. The beauty of the model is that each additional contributor further increases the data depth and usefulness, making the intermediary inherently more valuable to existing customers with every new customer acquired, and in so doing increasing the value of the platform as a whole.

Truly innovative models such as those cited here can powerfully disrupt or revolutionise many industries. But the financial services industry, benefits from these models particularly well; because at their core, financial products are based on information and participants may be reluctant to share their proprietary data sets with each other directly (as seen, for example, by the proliferation of dark pools). In order to make accurate investment and capital allocation decisions, most financial institutions materially benefit from knowing a consensus view and having a market reference price. Markit pioneered this give-to-get concept in the CDS pricing segment, creating more transparency available to all participants, building a large company along the way.

Entrepreneurs seizing the opportunity.

Serial entrepreneurs Donal Smith and Mark Faulkner know and understand the power of the ‘give to get’ concept as well as anyone. Having applied the contributory model so successfully to securities lending data in their previous business, Data Explorers, they're now transferring it to the rather larger opportunity of credit risk assessment. Today, we are excited to announce that we are leading a $7M Series A investment in their new venture Credit Benchmark, which collects, anonymises and aggregates credit risk data from the world’s leading global banks to create consensus credit risk estimates.

Credit Benchmark’s business model is built on a hitherto unleveraged resource – the data already produced by the credit divisions of banks and financial institutions for risk management and regulatory reasons. There are two reasons that these banks employ an estimated three times as many credit analysts as the credit risk agencies – a market which has long been dominated by just three players: Standard & Poor’s, Moody’s and Fitch.

First, they’re mandated by the regulators to have a certain amount of capital on their balance sheets, according to their exposure to different risk-rated assets. The analysts gather this information from estimates produced by their own credit risk models, which are regularly checked and validated by the regulators for accuracy. Second, given the recognised accuracy and up-to-date nature of these, they’re used by various other divisions to help analyse trading positions or risk. These are the true credit risk estimates – the validated views of qualified market participants with real skin in the game.

Credit Benchmark seeks to unlock the power of this untapped data asset by acting as a trusted arm’s-length intermediary, enabling anonymised aggregates of the data to be made available to other market participants whilst guaranteeing the confidentiality of individual contributors’ data. By playing this role, the company boosts transparency in the credit risk information market and allows every participant to work from the same dataset, just as there is one share price for a particular exchange-traded security at a particular point in time. Banks are now able to assess assets on their balance sheet with access to more information and can capitalise themselves in the most optimal way.

In Data Explorers’ final days they had over 300 customers contributing to the platform, highlighting the value that they took back from it. We believe banks will take little convincing this time. In terms of tech entrepreneurs, it would be hard to identify two more experienced and talented founders than Donal and Mark, who together with CEO Elly Hardwick, have the heavyweight clout and proven track record as a team required to have convinced a dozen of the world’s top 20 banks in the UK, Continental Europe and the US, to commit to providing data – and that’s just at launch.

It’s an absolute pleasure for Index to be working with the Credit Benchmark team, who we hope will be the first of many information services companies that we come to invest in.