Hedgehogs and Foxes

February 9th, 2021
by Mike Lines

I thought I would kick off the LIKEZERO blog with an idea based on some spurious philosophy, originally attributed to the Greek poet Archilochus and then popularised by Isaiah Berlin in his essay “The Hedgehog and the Fox”.  “The main principle being that the difference between foxes and hedgehogs, is that foxes know many small things, while hedgehogs know only one big thing.” He probably wasn’t expecting the idea to be evoked as testimony in support of a technology debate raging across the banking community, in fact Berlin said “I never meant it very seriously. I meant it as a kind of enjoyable intellectual game”.

So, what have hedgehogs and foxes got to do with LIKEZERO?  In 2011 I was a self-confessed fox, running a technology project to decipher complex contracts with a myriad of rules (knowing many small things).  The team spent months writing and refining rules to get good at understanding a single type of contract and we got pretty good at it, by 2013 we were still polishing these rules and were improved but were some way from perfection.  The truth is that however many rules you write there will always be bad results (false positives) in your data sets, so you must check everything.  And when you check results and complete missing data you must do this one document at a time, both handicaps challenge all “foxes”.

Sometime in 2017, the team and I had a revelation and now we are hedgehogs (hedgehogs know one big thing, which infuriates foxes).  Our single big idea was that rather than trying to spot differences in contracts with rules, to capture data, which is how machine learning technologies work, we would stand the problem on its head and match things that are the same.  In our initial prototyping we had some startling results, which gave us the confidence to invest.  As we started to acquire clients, we found that data mining and matching can solve the important and unsolved challenges in contract analysis: complex table structures, linking content in and across documents, multiple counterparties and amendment processing.

We think this is the way the market will now solve the problem of understanding unstructured data and make the journey to smart content and contacts.  It remains to be seen whether we will maintain our market lead in this approach – but we are busy little hedgehogs and we are confident that we will frustrate those hungry foxes!

OTC Derivatives UMR (Uncleared Margin Rules)

Following the first four phases of the Uncleared Margin Rules (UMR) for OTC Derivatives and the delays and disruption felt in 2020 due to COVID-19, it’s becoming increasingly clear how the regulations will start to impact a vast and varied population of market participants. This includes large institutional investors such as hedge funds, insurance companies, pension funds and corporates etc.

Implementing UMR not only puts a strain on legal teams, operations and fund administrators already stretched and overrun by overlapping regulations. It also adds commercial and reputational risk since trading can’t continue until firms have proved the data is fully operational. 

We’re ready for these key challenges and here’s how we can help you solve them.

Key challenges 

  • Phase 5 and 6 introduces a large number of buy side firms, some of whom may have limited resources (legal and data). This will also greatly increase the data challenge – the consolidation of fund threshold amounts across multiple dealers being just one issue.
  • During the hand-off from the negotiators to data operations, firms will need to manage data arriving in a variety of formats (PDF, Doc Gen feed etc.) as well as ensuring agreement data correlates to the correct counterparty relationship.
  • There are now a range of utilised options for negotiation, institutions that negotiate outside of the standard “playbook” will quickly erode any benefit provided and there will still be a need to extract the underlying data to feed internal systems.
  • Lack of consistency between industry data models and inconsistent data flows from negotiation providers.
  • Lack of detailed information and context in existing developed data models leading to a lack of confidence in the data extracted. 
  • No relationship between the collateral data and the legal text held in the document, make it a costly exercise to resolve disputes.
  • Mining useful data from Collateral Schedules that can be directly fed into a collateral system..

Contact us to find out how we are addressing these challenges.

LIBOR Cessation

Labelled the world’s most important number upon which trillions of dollars are resting in millions of financial contracts, clients are using our technology to support their LIBOR transition programmes to RFRs (RiskFree Rates).

With so many contracts impacted, it’s simply not feasible for any large business to perform a manual review exercise and the transition is not as simple as ‘find and replace’. Contracting parties need to understand current payment term mechanisms, hedging obligations, the parties responsible and any termination rights etc.

Here are some of the challenges facing businesses.

Key challenges 

  • The quality of historic record keeping will vary based on the underlying asset class. Due to various regulatory reforms, products such as derivatives would generally have good quality metadata, however, other asset classes such as Loans will require a considerable effort to load the “machine” with useful data.
  • Due to the fragmented phase out approach across different LIBOR currencies, determining the prioritisation of which contracts need review, and more importantly – which data is required, can be a big challenge and often requires a contract to be reviewed multiple times.
  • The volume and range of documents will mean that technology has to be agile and scalable for the ‘back book’ project but also suitable for BAU & the ‘forward book’.
  • Understanding your role in the contract – are you a “receiver” of a rate or do you have influence in amending the contract?
  • The complexity of the data model that needs to be fulfilled to allow for an easier remediation strategy without further need of more contract analysis.
  • Understanding the context of the LIBOR reference beyond a simple “LIBOR” search (e.g. is it referenced as an Interest payment).
  • Understanding the relationship between contracts where products are hedged.

Contact us to find out how we are addressing these challenges.

How smart

The answers our competitors are still looking for

Capturing data from heavily negotiated contracts without the need for human review or quality assurance is a challenge that many businesses face. We solve this by matching data (text strings) in a relational database – it means we never guess the answer, and once we know, we never forget.


A unique algorithm that brings everything together

A unique algorithm that increases efficiency / Time to end ‘one doc at a time’
For our competitors, once automated data extraction has run the process of completing and checking, the data is a painstaking job – one document at a time. LIKEZERO minimises the need for review but where this is required our unique clustering algorithm means that all human interventions are always on a multiple tread (many documents at a time).

We’ve got just the person for the job / The right people at the right time

Ensuring that an appropriately skilled person is used for every step of the process, is very hard to achieve. If this fails, the impacts are extremely costly. LIKEZERO both minimises the effort and orchestrates the work so that we always have the right people doing the best job. Lawyers deal with the really bespoke language whist operations staff can perform less skilled work.

Changing the model to work for you

Adding data to an existing data set is quite often more painful than starting from scratch. Our data framework allows you to load an existing structured data set and add to this so that you benefit from prior investments. You can also add data at any time to the data model at minimal cost or time.

We know the language of contracts / Interpreting the language of contracts

Connecting the dots in contracts / Managing the relationship of data in contracts
Managing the relationship between data in two different documents or in two different parts of the same document is a prerequisite for contract analysis, but it’s not easy. LIKEZERO has the capability to manage definitions and references to solve this problem, ensuring that you always know how to interpret the language in your contracts.