Enhancement of Treasury information workflows at RTL Group:
Data integration and valuation automation, continuous monitoring,
Improvement of internal controls and intelligent dashboarding
In collaboration with FinMetrics
The sophistication of IT systems results in a profusion of data and makes it very difficult
to process or compile it properly. We have to cobble dashboards together in a Heath Robinson manner because the data does not match standard formats and is compartmentalised in its own systems, which are all independent from each other and different.
Since too much is often worse than too little, we may end up submerged in data, or we may just underuse it. This is also due to another phenomenon: the preference of obscurantism to learning from imperfect but still adequate data, leading to Big Unused Financial Data. The Business Intelligence (BI) challenge is to extract the data, make formats compatible and process the data to condense it into an appropriate dashboard that makes sense for the financial managers of the Treasury. The added value lingers therefore on the whole chain, not on a complex data system, a nice visualisation tool, or a financial library, alone.
Excel: Not the best,
but still the most used TMS
For the last thirty years, Excel has been the piece of software in widest use amongst treasurers, but by far not the safest when it comes to use it at full throttle. It is not as robust as it needs to be, which is the other side of the coin, of extreme flexibility. However, for prototyping your next report or dashboard, Excel is at the top of the list, the one you think of immediately. It is capable of complex calculations and routines, and can communicate today with other databases of various dimensions. Unlike a Treasury Management System (TMS), Excel starts from a blank page and is not process-oriented. It is useful, but in measured, homeopathic doses. As with drugs, abusing from them makes us hyper-dependent. If CIOs are wary of new and untried applications, the temptation to use Excel is enormous. Here again, DIY can help us, but it will never be like a factory finish.
With a greater taxonomy of financial information and structuring it using digital tags, even with IFRS standards we will be able to extract financial data in XBRL type formats more easily. In this way data customisation can be easily parameterised to produce tailored reports, cutting costs. It is, and always will be, all about standards. One day, we will get there – but in the meanwhile we have to extract and keep on doing a bit of DIY. We may dream of a world in which a robot can extract everything and can discover by artificial intelligence alone how to compile it to produce intelligent reports for the management. Happily, or sadly, that will not happen any time soon, for two reasons. First, the lack of standardization decreases but still forces us to innovate and to sculpt a solution that can be swiftly parameterised to gather and integrate any data flow. Thanks to event-driven robots, we are able to do it more and more intelligently. Second, a tool remains a tool. Many of them remain a Lego box until somebody provides a shape to them. It is particularly difficult to have a programmatically efficient solution whilst being able to remain creative and capable to integrate the experience of senior managers. And creativity is not yet artificially provided…unless you believe in creationism.
The FinBoard Project at RTL Group
Why on earth would you look elsewhere for a solution and add yet another tool that has to be connected and interfaced? On the face of it, that would seem to be a pretty bad idea. However, the reality of current treasury IT tools is that they are standard, sophisticated yes, but somewhat generic and at the end, not very flexible. They are made to satisfy the greatest number and therefore they are off-the-peg products, when perhaps tailor-made would be better (except for the price). The requirements are becoming ever more specific in terms of reporting, and everyone addresses them in their own way. A generic solution is often not possible. Furthermore, the growth of regulations and new IFRS standards is such that software producers are a bit behind the curve and wait to find out what they have to develop before getting down to work. The principle of the “very last minute” (to put it kindly) therefore applies and very often leaves treasurers helpless in the face of compliance requirements.
Digressing a little bit on the question of “interconnected” vs. “integrated” systems, this has been a long and cyclical debate for a while. At the end of the 90s, substantial amounts were being invested by global firms in their desire to have one system for the whole group. Some of these projects were eventually trashed down, with huge losses. A better idea emerged at the same time: make systems collaborate more easily by being interconnectable (through connectors, APIs or the like) and by sharing a common taxonomy, the latter making connectors even more standard. Of course, pushing this idea to the extreme would be also critical, requiring at least more than one to know the map of interconnected systems. But globally, this simply means the mastery should not be on the system(s) but on the flows of information. Big system or not, few users are aware of where the data comes from…and even where it goes sometimes.
At RTL, like in many other corporate treasury departments, a wide range of specific operations were maintained in spreadsheets aside of those processes naturally performed on ERP and TMS systems. Those processes mainly comprise reports and internal controls, but also include the explanation of variations on the pricing components of derivatives used in hedging strategies.
Those checks and reports implied manipulations of large datasets, present on the main system, to be then aggregated with data flows from other systems in place or from external IT sources. Checks comprise verifications and reconciliations whilst reports require also mark-to-model valuations and specific splits of those valuations. The numerous manipulations and the repetition of cumbersome calculations on a regular basis – such as in FX where triangulated calculations do not lead to similar valuations – require an additional scrutiny as well.
This leads to many issues:
Ungrateful tasks making spreadsheets more error-prone and globally, creating demotivation and more human turnover.
There is a lack of responsiveness when gaps show up, when checks have to be done or simply when a question raised must be swiftly answered. It is perceived as a heavy process that must be performed on top of the rest. It takes time to reproduce those reports. By the time something is spotted, deadlines are often passed.
Productivity blackholes. Substantial energy and efforts are devoted on the replication of regular reports, producing values and cross-checking them, instead of having them automatically available on a regular basis and using them for timely decision-making.
Lack of global view on the processes run by treasury. The rhythm of the activity is regulated by the reporting frequency defined in the various systems. We want instead today an App to alert us on an issue raised.
The management has a very limited and irregular view on the activity and deadlines of the team. Still, spreadsheets have their role: they enable the team to prototype new functions, verify calculations and design new reports. The solution must be therefore one that automates the data
aggregation, performs financial and risk calculations in real time, and is capable to show an ergonomic monitoring dashboard that provides all reports and checks continuously, with the capacity to easily get reports and spreadsheet exports for the team for their own use and flexibility, but now at a later stage where cumbersome manipulations are behind. The champion users must keep ownership of the data.
In that respect, FinBoard (developed by FinMetrics) was the perfect application for RTL. It can be a full cloud solution only or combine a system on premises with a cloud interface, so that the team can perform very customised functions in-house, whilst publishing a complete set of key indicators to the App in the cloud so that the management can have a responsive cockpit on their desktop or smartphone, install-free.
Here are some key elements of the added-value by the solution:
Innovation and excellence. It provides an intelligent bridge between the main systems in place and the specific reporting requirements realized in spreadsheets in general. The tool collects data from different systems used by RTL treasury and store them in its own database. Several processes and reports usually produced using a mix of spreadsheets were transposed into the system, enabling significant savings in man-days as well as safer and more trustable results. KPIs and their drilldown, associated or not to market data from the usual providers, are also available directly on managers’ mobile devices. The solution converts reporting into monitoring.
The tool uses state-of-the-art technology guaranteeing a stable and robust system for the future. The data architecture, traceability and encryption have been designed by one of the pioneers of the Blockchain technology, recognized by Satoshi Nakamoto, Xavier Serret. The implemented solution can accumulate flat files, spreadsheets and data from APIs, maintain them in their original forms, with bots translating the data automatically for the various Apps. The overall chain is encrypted certifying the accuracy and integrity of the outputs as stemming from the same inputs.
Benefits to the users. FinBoard is not (just) a BI tool but a sum of benefits: (1) it makes robust and sound processes that are only realizable in spreadsheets otherwise, in a very short implementation time (days, max some weeks), (2) it provides a wonderful user-experience; it is ergonomic and ludic to use, fostering a frequent use, (3) it comes with a powerful instruments pricing and risk aluation library, (4) it provides Apps, each one corresponding precisely to a given function of the métier, and (5) it benefits from the accompaniment of experienced financial engineers who propose, design and can accompany treasurers in their follow-up of regulations and internal monitoring requirements. The savings in time and efforts but also the dialogue with responsive financial experts is the success key of the formula.
Examples of monitoring uses include:
Automation of EMIR Reporting process
Mark-to-market calculations and split between the value components (time effect, market effect, hedging impact),
Margins for back-to-back FX forwards with subsidiaries,
Process linking FX bank deals with associated FX internal deals to assess the split on/off balance sheet,
Wide KPIs on the overall activity as well as with data reused from other systems, spreadsheets, flat files, or live market data,The tool makes this directly available on manager’s mobile devices through an ergonomic cockpit.
An application to support a best practice in corporate treasury. By implementing an agile approach, whereby the Finance Engineers of FinMetrics accompany RTL team from the inception of the idea to the release in production, these ideas are quickly transformed into prototypes and then implemented into FinBoard as a new App. The Finance Engineers are finance specialists with an IT expertise, which simplifies greatly the dialogue between them and RTL Analysts. This leads to an immediate trust in the results and the “one version of the truth” produced. It means that the RTL team can spend more time in analysis and strategic tasks than in crunching numbers.