Founder member of FECMA

Advancing the credit profession

How to Drive Efficiency and Value with Artificial Intelligence – A guest blog by Serrala

Nils Strachanowski, VP Product for Cash Application, discusses how humans and Artificial Intelligence can collaborate in Accounts Receivable Management to achieve the best possible results.

Cash collection and allocation processes are more critical than ever in times of trade volatility and economic instability. What are the primary issues that businesses confront when it comes to cash allocation, and how are these challenges accelerated by crises like the recent pandemic, rising inflation, or continuous supply chain disruptions in the UK and globally?
A series of difficulties that we find in every company that impact the cash allocation process includes manual and error-prone processes, limited knowledge transferability, outdated technology, or lack of cash visibility. Primarily, we have manual and repetitive processes that slow down the allocation process and cash flow – this is a big issue, especially when liquidity is limited, and you need your cash in the business as quickly as possible. Sluggish processes are frequently the cause of delayed cash allocation. Many organisations also suffer from a lack of cash visibility, which may be especially problematic when dealing with a crisis and needing to know where your money is instantly. Furthermore, finance departments are unable to fully utilise the potential of their cash allocation due to fragmented processes and limitations of currently implemented solutions.

How can you assist your clients in meeting and overcoming these obstacles?
Manual and repetitive tasks may be easily solved by intelligent automation. Finance functions may reach automation rates of nearly 100 per cent in cash allocation using relevant technology and decreasing error-prone human intervention across the board. Technologies as such help with the monitoring of crucial KPIs in O2C and obtaining complete transparency of the process by reducing unallocated cash significantly. Organisations also benefit from a higher degree of standardisation if they want to analyse all incoming formats automatically and use intelligent technologies like artificial intelligence (AI) and machine learning (ML) to speed up cash allocation and improve cash flow.

What does a best-practice automated cash allocation process look like?
Cognitive cash collection can generally be divided into 3 categories. Firstly there is ‘neat end to end processing’. Standard forms – such as bank statements, remittance advices, settlement files etc – these need to be acceptable in all client and supplier formats. Transparency is essential and can be assured by ensuring pre built KPI’s and KYC elements are included – there should also be capability for individual users to create matching criteria and integrated workflows. Secondly, technology should be ‘future proof’ – combining AI and machine learning for best in class solutions and Third; they should eliminate or automate repetitive processes – allowing focus to be placed on value-add activities.

After having implemented such technology, organisations see high automation rates and can eliminate manual and repetitive work by an average of 85 percent. The result: same-day receivables matching, no more unallocated cash, lower DSO, and more efficient order-to-cash operations.

 Artificial intelligence and machine learning were mentioned. Because of their self-learning and self-acting nature, they are also referred to as “cognitive technology.” Can you give an example of how cognitive cash allocation works in practice and how this technology benefits businesses?
Key is to have people and AI collaborate to get the best results possible. Even for complicated cases, cognitive finance systems provide people with rules that are straightforward to set up. You may use any information from bank statements, remittance advices, settlement, and lockbox files to quickly change pre-configured rule templates.

Even without any defined rules, AI can easily match receivables. Client payment behaviour is identified, which influences credit scoring and decision-making, and machine learning is used to learn customer names and bank account numbers. Automatic learning is also being utilised to enhance the identification of remittance advice.

Finally, cognitive technology can quickly detect exceptions, allowing for faster handling of bank statement and lockbox exceptions. Automated posting suggestions speed up the exception management process, including the automatic initiation of follow-up activities.

Do you think cognitive technology will play a significant role in Order-to-Cash in the future, based on your experience and feedback from your clients?
We can see things heading in that direction. In recent years, remote working has stressed the significance of a strong digital infrastructure once again. As a result, finance digital transformation has been bumped again on the finance leader’s agenda, and an end-to-end approach is critical to ensure success. When starting a cash application automation project it’s worth considering a 360 degree approach for even improved results, adding automation solutions for payment requests and the reconciliation of settlement files from payment service providers. They are a fantastic match for the cash allocation process, and they are yet another example of the end-to-end automation strategy.


Nils Strachanowski, VP Product for Cash Application, Serrala
As VP Product, Nils is responsible for ensuring the success of the order-to-cash solutions, particularly in the area of cash application, at Serrala. Prior to this position he worked as Senior Solution Architect for Serrala. To date, he has supported over 100 customers and has implemented multiple successful automation projects across the entire EMEA region. He regularly writes articles for finance and treasury magazines and is an experienced speaker in the O2C area.


comments powered by Disqus