Factoring is a business that can only be competitive and successful with a high degree of automation. For this reason, factoring companies have always been on the lookout for new ways to automate their business processing.
Against this backdrop, it is hardly surprising that factoring companies are also looking forward to the exciting new opportunities offered by artificial intelligence (AI) in view of the triumphant advance of artificial intelligence across a broad front. Conferences focusing on AI are being organized, courses are being booked and various AI tools are being experimented with.
Hardly anyone will doubt that a certain amount of hype has arisen around the topic of AI. Many companies are almost obsessively trying to use the label "AI" - and often it is only simple sets of rules that are referred to in this way. This brings the use of "AI" close to the greenwashing that has crept into the topic of sustainability in recent years and turns it into AI-washing.
This entails the risk that the often superficial benefits of the "AI label" could distract from the actual objectives: Making the business more efficient. This means more transparency in the preparation and processing of transactions as well as better decision-making and risk control during the business relationship. AI should be a (further) means to this end rather than an end in itself.
In this article, you will find out what the current reality really looks like: What can AI currently do in factoring? What can be expected in the future - especially against the backdrop of an increasingly strict regulatory framework for the factoring business? We would also like to show you alternatives with which efficiency gains can currently be achieved faster and more extensively than with AI.
Analysis of AI in the factoring business: its role today and in the future
Overview
In order to work out the role of AI in the factoring business in more detail, we examine the following four fields of action:
- Onboarding of factoring clients: This is about the process of onboarding new factoring clients (follow-up clients). This traditionally includes the economic assessment, from obtaining credit information to a comprehensive balance sheet analysis and visual assessment ("eye diagnosis"). In addition, the business model and customers (debtors) of the follow-up customer are examined in order to make informed decisions about the purchase of individual receivables or the definition of credit lines. This applies to the various factoring products, such as genuine and non-genuine factoring, full service and open or silent variants.
- Communication with customers: During the business relationship, close communication between the factoring company (factoring provider) and its factoring clients (follow-up clients / factoring recipients) is crucial. This concerns all issues relating to the business relationship and the current use of services, such as receivables submitted or purchased.
- Purchase of receivables: Typical components of this area of activity are the automated transmission and verification in relation to the purchase conditions in accordance with the factoring agreement - in terms of content, scope (limit check) and quality (protection via trade credit insurance). Increasingly, the focus here is also on (extended) checks for early fraud detection (fraud prevention).
- Contract management: Execution of all business transactions after the purchase of receivables, starting with repayment (automated processing of incoming payments) or defaults (reminders or forwarding to debt collection agencies).
Particularly in hype phases, in which technologies such as AI are currently seen as a "wonder weapon", strict practical relevance is especially important. As a specialized service provider, we are active in the relevant forums and associations, closely follow the developments of AI providers and attach great importance to practicability, feasibility and usability - always with a view to the requirements of factoring companies. This determines the subsequent assessment of the current situation and encourages us to forecast the situation in 3 - 5 years' time - knowing full well that modern technologies are developing (almost) exponentially and their impact is therefore usually even underestimated.
The role of AI in the onboarding of factoring clients
Current:
AI is already being used regularly in some places to improve the onboarding process. The focus is on gathering information about the factoring client, their economic situation, their specific business and their business risks. A lot of information can be gathered (faster and more systematically) independently of an "eye diagnosis" and on-site assessment. The workload of employees is reduced, while risk management and portfolio management can be supported by simulations. AI technologies are therefore already helping to make the process faster, more efficient and more customer-specific.
The future:
It can be assumed that AI will increasingly penetrate the area of onboarding in 3 - 5 years (or even earlier). Through advanced automation and deeper personalization, AI will be able to autonomously control the entire onboarding process from data collection to risk analysis. Proactive problem solving will be performed by AI systems by default, reducing human involvement or focusing on the skills of assessing the information at hand. However, as the human component should continue to play an important role in maintaining long-term business relationships in the future, AI will not completely dominate this area.
AI as a helper in communication with factoring clients
Current:
The use of AI in communication with factoring clients, particularly via client portals, is still at an early stage. Although there are already initial applications, such as AI-supported chatbots and special "sets of rules" for documents and papers to be submitted, the comprehensive use of AI in this area is not yet the norm.
Future:
AI will make it possible to take communication with factoring clients, as well as their debtors, to a new level by taking into account client status, behavior, business volume and evaluation needs. AI-supported chatbots and virtual assistants will be firmly integrated into most customer portals to make communication more efficient and personalized. AI will prepare the interpretation of submitted documents "as a particularly diligent assistant" and contribute to the decision-making process.
AI as a pacemaker in the purchase of receivables
Current:
AI is already being used successfully to automate invoice processing and in risk management. These technologies are increasingly finding their way into modern factoring systems and can use large amounts of data to supplement or correct missing information, detect irregularities or anomalies and optimize the utilization of credit limits. AI can also automatically initiate (standard) communication with the customer or make appropriate suggestions to make processes more efficient.
The future:
AI will play a dominant role in receivables purchasing by (further) automating the entire invoice processing and dynamically managing risks - especially in relation to the current growing need for fraud detection and prevention. The integration of AI with blockchain technologies will lead to even more secure and efficient processes, across companies in the sense of an embedded finance approach provided by the factoring company to its factoring clients. This reduces human intervention to a minimum and the purchase of receivables can run largely autonomously. Alternatively, the process could be subjected to additional AI-supported checks to ensure even greater accuracy and security.
The functions made available to debtors are also enabled by AI and can therefore fulfill their information requirements better and more dynamically. A major change will result from the (improved) ability to provide targeted information, with a swing from "pull" to "push".
AI as an efficiency lever in contract management
Currently:
AI has already gained a foothold in contract management, particularly through the automation of routine tasks and the monitoring of ongoing business activities. AI enables a higher degree of automation in the allocation of incoming payments, even if the payment information is incomplete. The use of AI opens up considerable potential, but many processes still require human monitoring and control - at least until appropriate AI learning curves can fully take over these tasks.
The future:
In the next few years, AI will be just as dominant in contract management as it is in receivables purchasing. AI will be able to manage contracts autonomously, including proactive risk monitoring and dynamic adjustments, particularly in the event of defaults. AI will also "know" when and how to approach which customer for outstanding payments - and thus optimize the dunning process.
In addition, AI will be used in all internal processes that create transparency, enable control and generate reports in order to meet the growing regulatory requirements. However, human input will continue to be required - especially for more complex business constellations or special legal requirements (in international business); corresponding resources no longer need to be tied up in routine activities due to a shortage of skilled workers.
Summarizing overview
In 3 - 5 years (or less), AI will predominate in all areas of the factoring business. While the human component will continue to play a role in onboarding and customer communication in the future, the penetration of AI in the areas of receivables purchasing and contract management will go one step further: AI will achieve an absolutely dominant presence in these two fields of activity.
This reflects the generally advancing development and integration of AI in business processes for the factoring business as well - resulting in an even more efficient, secure way of working with a higher degree of automation.
Potential impediments / obstacles to AI
You don't have to be a great prophet to determine the future role of such a powerful topic as AI. If technological progress continues at the speed at which it has begun, AI will soon dominate in automatable processes in the factoring business too.
It is therefore particularly important to consider opposing trends. The topic of AI actually thrives on decisions that are not transparent (for humans) based on independently and freely collected data. However, every factoring company knows that legislation demands exactly the opposite - and increasingly so: transparency regarding the basis for decision-making. For this reason, here is a look at the most important opposing forces that could limit the cheering forecasts for AI in the factoring business.
Regulatory restrictions
Compliance and supervision: Factoring companies must adhere to strict regulatory requirements, particularly in relation to financial transactions and the prevention of money laundering. AI models must be developed and deployed in such a way that they do not violate these regulations. As regulatory requirements are often very specific and complex, the integration of AI can be challenging as every automated decision must be traceable and compliant. This often requires detailed documentation and monitoring, which can limit the flexibility and efficiency of AI.
Transparency and explainability: Regulatory authorities require decisions, especially in the financial sector, to be transparent and understandable. AI models, especially those based on deep learning, must be seen as "black boxes", as their decision-making processes are often difficult to understand and the corresponding decisions cannot be quickly justified. This lack of transparency can also lead to regulatory challenges, as companies may not be able to fully explain the AI's decision-making processes.
Data protection and data security
Data processing and storage: AI models require large amounts of data to work effectively. However, in the factoring business, much of this data is sensitive as it contains personal information of companies and their customers. Data protection laws such as the GDPR in Europe restrict how this data may be collected, stored and processed. Among other things, companies are required to obtain the consent of data subjects and there are strict rules on storing and sharing this data.
Data availability and quality: AI depends on the availability and quality of data. In the factoring business, however, sufficient data of the required quality is not always available, e.g. smaller factoring companies "only" have their own database available, which may not be representative. The need to keep data anonymous or the need to expand your own data can impair the performance of AI models. In addition, the linking and exchange of data between different systems or organizations can be significantly restricted by data protection regulations.
Security requirements: Data in factoring is often of critical importance and the security of this data is a top priority. AI modules, especially those based on cloud technologies, must meet strict security standards to prevent data breaches, as do systems operating in the cloud in general. This can slow down the implementation of AI and increase costs.
Ethics and accountability
Ethics of AI use: The use of AI also raises ethical issues, particularly in relation to automated decision-making. It must be ensured that AI systems work fairly and impartially and do not make discriminatory decisions. These ethical considerations can also slow down the development and use of AI systems in factoring, as extensive testing and monitoring is required.
Accountability: When AI-supported systems make decisions, the question of accountability arises. Who is responsible if an AI decision leads to an error or damage? This question is particularly important in the highly regulated financial sector and can make the use of AI complicated. In particular, the aspect of "hallucinations" should be taken into account, which is currently (still) often observed: the statement of an AI model sounds plausible, but on closer inspection it is not. As the particularly hard-working assistant, AI should therefore be looked over the shoulder intensively for a while until it is allowed to "run on its own" in suitable business processes - in the spirit of Goethe's sorcerer's apprentice.
Conclusion: What is currently the best way to deal with AI in factoring?
AI in factoring can already be seen in all areas of activity. Technological suitability and progress also suggest that AI is likely to dominate some process steps in the coming years. This will happen faster than we can imagine: With each new release of the innovative software providers (and in in-house development), we will recognize another building block. However, the force of the sometimes opposing, obstructive currents such as regulation and data protection must be taken very seriously. In fact, we believe they are so strong that they could lead to increased expenditure in order to make AI effective in many areas of factoring over the next (5) years.
This uncertainty is exactly what we argued at the beginning of this article. Despite all optimism, it would be ill-advised at this stage to "jump the gun" on the topic of AI in factoring. Instead of spending a lot of energy on initial applications and arguing that AI is in use, as is the case with greenwashing, we should examine how automation can be driven forward without AI.
In practice, there are many automation options in the factoring business. The performance of modern IT solutions often reveals potential and is gradually bringing AI models into use in their standard development. Our HENRI for Factoring solution is an illustrative example, as it is based on Microsoft Dynamics 365 Business Central and therefore benefits from all the developments of one of the major hyper scalers. HENRI for Factoring is tried and tested, ready to use and can be individually configured to meet specific customer requirements, products and processes.
