Interview: Automation of factoring services for SMEs.

As Head of Sales, Daniel Huszár knows exactly how international markets work.

How can or should banks automate? What might the exact process be?

First of all, if we go into the micro sector and SME sector with a manual process, we don’t get so much more money from each individual client as for example in the corporate world. So, yes, there is really a need to automate. With manual processes versus automation – automation makes things scalable, which is very important with the SME and micro-segment. Because we have a lot of customers and because it’s hard if we have manual processes connected to all of them, when we onboard them and manage them, we need to automate, because then it’s not scalable. And perhaps when we are dealing with corporates, in the case of invoice discounting or factoring products, we don’t have a large number of clients but we do have a large number of invoices, and we kind of need to automate in order to get them into the system, to manage the limits and to manage the risks.


So, we need automation, and to do this we have two technologies: we have the front-end, which is an online portal or an app, something like that, and this is the face of the technology that the customer sees, and then we have the back-end, which is where you crunch all those numbers, make all those postings, track the invoices and track your risks, for example. I’ll start with the front-end and what we can do there. I think, with front-end, the design is the most important thing, and it’s not really about automation here but getting the customer onboard so that they want to interact with your solution and so that they really want to use it, because without that, there is no product to use. To do this, we definitely need to shape it from the mobile, and by this I mean make it really simple to use, make it easy to upload invoices or documents and to make requests and all those kinds of things, and give users multiple opportunities to interact with your technology and your product. And when the automation comes into play, it’s more about customer convenience.


When I talk about a multichannel approach and multiple ways to interact with your software, it’s more about ‘do I type in this invoice manually or do I take a picture and then get some OCR software to read out what is written on that picture and translate it into the fields?’ Customer convenience definitely plays a more important role here than even automation. It just needs to be easy and simple. With respect to the back-end, the most important thing is flexibility. It has to be possible for all of your business processes to be translated into the software’s parameters, which are really the rules by which things should be automated – for example, which invoices do you buy, which clients do you onboard according to severe ratings? So those are the most important parts.


What are the most important or crucial steps for banks or factoring companies when it comes to automation?

Firstly, I’d really urge every product manager that conceptualises their product to completely step away from the tools that they’re currently using, because they kind of limit the way we think, and the product shapes itself after the tools we use and it’s hard to get different ideas or see things from a different perspective. So, if you are making a new product for SMEs, let your imagination run wild. Not too wild of course . Think of everything, and then we can confine it to the product. First, of course, we need to see business cases, if feasible. This would be step 1 for me, to get this concept.


And step 2 would be to translate this into the technology, i.e. to make the front-end tool easy to use, to have this multichannel approach that allows the customer to interact with you on many different levels. For example, as I mentioned, this could be entering invoices using various methods: by taking a picture, typing them in. And the customer’s convenience is key. For the back-end, the most important thing is the way you automate things, the way you translate your business processes into the software parameters – that’s basically it. You take whatever rules you have in mind, like which invoices to buy, which customers to take, which customers to reject, how to set your limits, and you kind of think about what your risk appetite is here, because without risk we can’t do this. We have to take some spread, we have to take some risks, and we are not getting everything covered by credit insurance.


Do you understand automation to mean that all processes are done by machines or robots without any human touch or involvement?

This is an interesting question. I don’t think so. It’s not that it’s dangerous, but it doesn’t need to be that way. Of course, automation means that things are run automatically, but when we look at certain risks, if you have hundreds of thousands of customers or invoices, then nobody can look over all of these, and you need pre-defined decisions, but you need to look at some of them and you need to look at where risks are most likely to occur. This is what big data risk management is for, so you put it all in and you let the technology run its assessments according to some predefined rules, checking your invoices, checking your customers against patterns. For example, if you can look at the invoices and invoice numbers, and you can check if there’s a pattern that can be flagellant. So in the end, when we apply all these parameters to our data and we think there might be a risk, like the Bedford model, and patterns and all these kinds of things, then the risk manager will get a proposal list and, rather than look at every invoice or every client, they will get the best clients according to the business rules set within the software, and then they can see which clients or invoices need to be checked up on the most.


Basically, automation can also mean that humans use their time in the most efficient way possible. Automation should serve humans and users, not vice versa, of course. It’s just a tool. To make sure that it’s not a black box, I believe there are two factors that really help. I mentioned that we translate business processes into parameters so that the software, when you automate it according to the business rules that the bank and we as a software company implemented in the first place, does exactly what you want it to do. So it’s just a process enhancer: it takes business rules and applies them over and over, and quicker. But it is what we tell the software to do. And the second thing which is very important is that we need to be able to analyse all the data that goes in, and what we at efcom do to ensure that is save everything at invoice level, so every invoice and every credit note, every item that goes in there, stays within the database and can be analysed upon reporting even years later. That is really important when it comes to shaping your portfolio, monitoring it over time and even having the possibility of creating reports or dashboards later on, which are things that you might not think of right now, because they represent future products. And that’s a very important factor, in my opinion – that you are able to do that.