Cutting the huge costs of eTrading desks with tcapy

If you read my column, you’ll probably know I like burgers and make any excuse to mention them. Today is no exception! Let’s say that today, I wanted to open a burger joint. Ok, I’m no chef, but I think I can probably make a burger. Rather than simply buying a ready made building to house my burger joint, let’s say I insist in building it myself. Furthermore, I’d insist on making all the bricks. I’d also insist on building all the fridges, the cookers and every bit of equipment. What might have been a quick project would end up taking months, requiring lots of specialists. It would also likely cost a massive amount of money to build everything in this way from scratch. It is unlikely that doing all this work would differentiate my primary offering, the actual burger. I’d be better off spending my time improving my burger recipe!


The difficulty is that this type of approach of building stuff from scratch is something that happens often in financial firms. The result is that it can become very expensive to run desks whether on the buy side or sell side. Take for example an FX eTrading desk in a bank. The desk is likely to have spent millions developing all its analytics, whether that’s creating pricing engines, execution algos, TCA tools etc. There’s also the massive annual cost of maintenance. Of course, some of this code will be highly proprietary. However, the bulk of the code in, for example, a TCA tool will be quite generic and very similar to TCA tools across the rest of the banks, with some code on top to differentiate it. For any eTrading desk, it is a necessity to provide TCA tools for their clients. Banks are spending millions and millions of dollars across the street and lots of time on what is mostly generic TCA code, taking up time and money, which could be spent on the differentiating elements of TCA.


How can banks stop spending massive amounts of money to develop and maintain FX TCA platforms? Cuemacro’s tcapy is the solution, which will save money and time! It is written in Python, so it’s easier to maintain, rather than being written in difficult to read proprietary database query languages. Firms which purchase the tcapy library, get access to all the source code, so it can be fully embedded into their internal systems for trade and market data. For the first few clients, the licence is a one off cost, which is a relatively unique concept in the industry (most external TCA services come with a compulsory annual licence cost). An annual technical support package is also available as an option, where Cuemacro can help maintain the code, and you’ll get access to tcapy code updates. The cost of the support is significantly cheaper than maintaining your own internal TCA. Your internal team can spend time and effort adding customisations (or Cuemacro can be contracted to add new features) to differentiate what you do in TCA, rather than spending a lot of time and money maintaining the generic parts.


The costs of running eTrading desks are huge. Having generic building blocks for analytics like TCA can help massively cut costs. Furthermore, tcapy is fully transparent, and can be run internally, so trade data stays internal and private. Hence, you get the benefits of an internal solution too, whilst at the same time it’s an independently developed solution, which clients appreciate. If you’d be interested in hearing more about how tcapy can help cut the massive costs of developing and maintaining your own TCA library, let me know!