What was your remit when starting at the company and how has that developed in the last two years?
When we were designing Gresham Clareti Transaction Control (CTC), we identified a large gap in the traditional reconciliation space. Legacy reconciliation vendors have been around for many years tackling SWIFT cash and securities reconciliation, which they do well. But at Gresham we were very conscious that there was a whole set of new message formats coming out for trading more exotic instruments, and the complexities and the data structures around those message types were a lot more complex than SWIFT messages.
Firms also wanted to reconcile earlier on in the trade lifecycle, near the front and middle office. Because of the way in which those legacy systems were architected, it was difficult to incorporate the changes. To get around this there was a huge increase in the use of Excel spreadsheets as controls. That’s because they are quick, they’re flexible and anyone who is smart in the middle office can write a macro and start to compare data very quickly.
But the problem that comes with that is a real lack of control. At best, you can see who saved the spreadsheet last, and there are greater possibilities for fraud. This has been borne out in some of the more high profile cases involving big banks, where the errors were based on spreadsheet-type controls. So regulators are now starting to say spreadsheets can’t be used as an effective control.
What we wanted to do was design a new reconciliation tool that would allow firms to on-board the new, wide data structures, just as quickly as with Excel, but with all of those enterprise-wide controls that you need within a reconciliation system: scalability, robustness, integrity and security. We wanted to disrupt a very stale market that is dominated by only a handful of vendors and bring something new to the industry.
Compared to the legacy systems, what differentiates Gresham CTC?
One difference is our rapid on-boarding capabilities. We are renowned in the market for the speed in which we can on-board new reconciliations. We can do that at a fraction of the time of any of our competitors. It’s interesting to see in the last 12 months how those legacy vendors are trying to respond to us in terms of releasing new products. But they are only getting about halfway there.
A second difference is that we don’t use a fixed data model and specifically design systems to be flexible. In contrast, the legacy systems were built primarily to reconcile SWIFT messages based on fixed data. In order to adapt these systems you have to go through shoe horning exercises using expensive extract and transform data and technologies to get the data into a format that you can reconcile. Gresham CTC is not constrained by any fixed data models. If we want to reconcile a Depository Trading & Clearing Corporation message that has 700 attributes, we don’t have to make any compromises.
Because we’re using newer technology we can deploy much quicker and leverage new technologies to give us real performance and scale. We did some benchmarking tests with Intel where we were loading and matching 1.8 billion equity trades in one hour. That’s 500,000 per second. None of the legacy vendors come close to that sort of performance. For us, it is about rapid on-boarding, no fixed data model, and performance and scalability.
With the mandates you are winning, are you taking on legacy systems or are clients moving to yours?
For the majority, we complement those existing systems. There have only been a handful of deals where we have replaced them. We don’t suggest that you go and rip out these big systems, as people have invested in them for a long time and they do work well in cash and securities reconciliation. That is not where the risk lies. The risk lies in derivatives and new structured products that don’t fit into the legacy model, which are run on spreadsheets and touch many different systems. We find that most of the mandates are either brand new requirements or complement existing infrastructure.
Gresham also has an account receivables management solution. We’ve been working with one of the leading banks in Asia to design a specific solution for this problem, particularly around the bank-to-corporate margin. Banks want to be able to show stickier relationships with corporates, especially since liquidity capital ratio requirements are calculated on this basis.
We’ve built this management system and we’ve got some real interest in that from some of the big tier banks that are looking to buy the solution. Fundamentally, it is still reconciliation and matching, but it is specifically around accounts receivables management.
Banks are always trying to renew products, which invariably means there are always new systems. The whole world of data is crazy and you have regulators bringing in extra reporting with the US Dodd-Frank Act and European Market Infrastructure Regulation, so it’s not just about the checks and the balances, the reconciliations need to be able to help.
Since 2008, there has been a huge raft of regulation to try and make things safer for the customer. Would you say regulation has been a real driver in this industry?
Definitely. It’s the growing need to prove integrity. Regulation is one angle that is forcing banks and institutions to be accountable for their actions. A lot of that comes from social and political pressure, but if you get this stuff wrong now you can go to prison, or you can lose half of your wealth. The days of not worrying whose money it is are long gone now. So, the guy who signs off on whether an operation is good or not needs to prove that integrity, otherwise he may be doing a 10-year stretch on Breaker Island.
People are also looking for robustness and completeness across the front and back offices. A really good example of a piece of regulation that is having an effect is Dodd-Frank, which says you have got to report from a group-wide basis. Before, people would just report in their silos. The front office would have to report their profit and loss and their risk. The middle office would do their own reporting. And the back-office would also report separately. In order to improve integrity you now need overarching reporting and holistic controls.
Is the need for a holistic solution making banks address how many systems they use for each of the offices?
There have been lots of studies on this. The regulators are calling them user-developed applications (UDAs). What they mean by that are the spreadsheets, access databases and home built applications that do the checks and balances. The reason those systems were built in the first place is because legacy vendors can’t deal with some of the requirements. Over time, people have either got by with spreadsheets, which are all over the place, or the more sophisticated ones have built applications to handle the problems that the legacy providers don’t deal with.
I see banks go on these programmes and say, “we are going to cut down our systems from 300 to four”, and you think that makes sense. What happens is, they start writing these requirements, and five years down the line they deliver a small number, but it is not until nearly seven years after they started before they’re down to four.
The problem is the guy in the front office doesn’t want to wait. All of a sudden he’s saying, “I’m now going to trade X & Y and I need a system for that”. The front office always gets what it wants as it’s always innovating and trying to stay competitive. What starts off as a good idea is quickly out of date. By the time they have finished, it doesn’t work, or they have new products in the front-office they are trying to develop. So they go round in this cycle, burning billions of dollars in the process.
If there was one thing you could change in the industry, what would it be?
I would like to see quicker decision processes within the banks. I think they take too long to make a decision. I am not sure whether this is because they are extremely risk averse, or whether they don’t look that far ahead in time. The decision making process can drag on and can often be longer than the time it takes to implement the solution.
Next interview →
SmartStream
Darryl Twiggs