Data Procurement in Credit Management
“You’re a messie!”, that’s what my wife keeps saying to me when she visits my hobby cellar. Since I take many things into my own hands, I also need a lot of tools as well as the appropriate utensils or spare parts. She simply doesn’t understand that!
But isn’t she a little bit right? When it comes to computer technology, the motto is “just bought and already outdated”. With the good old Bosch hammer, on the other hand, the half-life is much longer. That’s why I always re-evaluate the things in my cellar through my knowledge and experience. And indeed, I sort out a lot of things. My wife would certainly sort out even more, she’s simply more the type to do so.
So would she be the better data manager?
According to the statistics portal “statista”, 16.1 zettabytes of data were generated worldwide in 2016. For the year 2025 “statista” forecasts a tenfold increase to 163 Zettabyte. A small excursus: One Zettabyte storage capacity stands for 1021 bytes. These are sextillion bytes or, in numbers, 1,000,000,000,000,000,000,000 bytes. This in turn corresponds to 1,000 exabytes or one billion terabytes.
This data big bang began in the mid-1990s. Incidentally, that was when Amazon was founded. Hardly anyone will really be surprised about these gigantic quantities any more, since more than half of the world’s population, i.e. about 3.8 billion people, are now active Internet users. The proportion of those who generate a particularly large amount of data in social media is 40 per cent.
But now back to my reassessed basement. And anyway, what does all this have to do with credit management?
Well, just as I have to decide in my basement for the sake of order what I still need and what I don’t need, everyone who handles data must regularly ask themselves: “Which data is important? Which not? Which do I still need to save and which can I even delete? Which other data is reliable? Which is unreliable? And most importantly: “Which data is even false or fake? In order to evaluate data, we all have to mix a suitable cocktail of data in order to draw our conclusions. I have to do this in my basement in exactly the same way as the credit manager who uses existing data to evaluate his customers.
There is internal and external data for evaluating customers. This is used to form key figures such as payment experiences, balance sheet total, turnover, probability of default, etc. For you, dear readers, this is not news. However, the increasing speed of data being changed is making the whole construct obsolete faster than it has been in the past. In addition, the data generated from the social media area is becoming more and more important and, above all, more and more abundant!
But be careful with too much attention to the social media data! The difficulty here lies above all in the evaluation. Also automation and the use of AI is still difficult to implement here, since it is not yet possible to reliably filter fake news (see the recording of our webinar: “That’s really all Fake News?”). – On the usefulness of social media monitoring in credit management). Is the data source even a bot or a mass of bots? However, social media data are already being used as so-called signal data (please also read the article on this topic regarding the use of alternative data sources in credit management).
Back to our data cocktail, which every data manager must prepare himself
When evaluating customers, for example, this can be done per country, per business unit, per product group, perhaps also per sales channel or even sales staff. Here, of course, principles such as commercial prudence must be observed and thus constantly threatening default risks minimised. For a new customer, it is also better to add a little security such as insurance or a guarantee such as a surety. In the course of time, payment experiences will develop and thus further ingredients for the cocktail.
In conclusion, apart from more frequent reviews of the data mix and thus of the valuation rules in classic credit management, nothing fundamental has changed yet. Due to the increasing importance of social media data, however, a change will take place here. The aim must therefore be to write a rule that clearly states which information is used in what quantity to obtain reliable customer ratings. For the sake of completeness, such a set of rules should be included in a credit policy. This should be reviewed and critically scrutinised more frequently than in the past. The old thought: “we’ve always done it this way” must be a definite fit and belongs in a museum! Regular, critical scrutiny of the rules and regulations must be part of the credit policy. That’s exactly what I will be doing in my cellar in the future.
How nice would it be if there were a kind of master control programme that allowed inhomogeneous ERP or CRM systems to communicate with each other, mastered the exchange of information with a credit agency, offered the possibility of including social media data and at the same time guaranteed compliance with the credit policy? All this on a platform and in a system that additionally draws the user’s attention to unusual behaviour of his customers in the form of daily routines! Already exists!