Data is the new Gold… What does that really mean for treasury and finance?
Steven Batiste, the CTO of TIS, explains that gold is something that people value, which is exactly how he sees data! Steven feels that data is not the new “oil”, because that would be like storing a potentially valuable commodity without refining it. Information is valuable because of what you can do with it, not just having or owning it.
There is a massive amount of information that can be gathered from sending and receiving payments data. Once this information is harvested, then analytics can be performed. Within the finance area, companies have a lot of good data allowing them to see patterns that can help in the decision-making process. These results can support conclusions that guide a company’s strategy, investment policy or support fraud and cybercrime detection and prevention.
Being in the “holistic” center of payments and financial information has significant advantages for data analysis… as well as implications. “If you are where it all comes together like we are at TIS, i.e., having the ability to centralize multibank and multisystem information, you are ahead of banks and other players that only see their share of the enterprise’s volume,” explains Steven.
Steven agrees that having data that is siloed can get tricky. Even with good AI, if the information is not accessible then there is no advantage. Unfortunately, there can be various barriers e.g., because we no longer know how to access it (!), there is no standardization in the formats or data can be off-limits due to regulations such as GDPR. Data lakes can often bridge various gaps and support appropriate consolidation, even segregating data that is restricted.
Technology has indeed become a game-changer e.g. through cloud access and storage, or AI and machine learning. However, beware… AI and machine learning are only valuable if you have clean data, in other words: garbage in / garbage out. Information needs proper and careful refining to become both useful and valuable. A data lake can become a data swamp when information is polluted.
Steven explains that it is important to get rid of biases, but that is often difficult. An example of potential bias is connected with the pre-screening of applicant profiles on LinkedIn. If the system focuses on one type of profile early on it can become “biased” and begin looking only for others that conform to this first example, thus building bias into the process.
What about regulation? Are the restrictions too great to make data truly valuable? Steven does feel that regulation makes it more difficult, but there are ways to anonymize and aggregate information so that it is usable and useful! However, the key is actually not the regulation itself, but the fact that each country or jurisdiction e.g., many states in the US, have their own brand or flavor of the regulation.
Steven feels it is certainly the big techs who are in the lead in regard to data. However, large and established banks also have a lot of good information going back for decades. The challenge is to understand how to get full access!
This of course begs the question, should you even bother to start collecting and analyzing data if you are not one of the big(est) players? Do you even stand a chance? Steven’s resounding answer is YES! It will always come down to how you refine and analyze your data, using patterns to help draw conclusions. The sophistication of your AI models and indeed a company’s products is also key. Without desirable and competitive services, even the best AI will not help you to grow revenue.
Data does not have to be as exclusive as it may seem at first glance to be valuable. There are many platforms where an enterprise can leverage its potential by gaining access via API to historical data or machine learning capabilities. There are pre-trained models that will let you use your own smaller collection of information and enrich this with someone else’s historical data to give you good and solid results. It’s never too late to start!