You no longer need vast amounts of data to automate accounting.

Automating transaction categorization has traditionally relied on conventional Machine Learning (what AI referred to before GenAI arrived) or heavy ERP machinery. The problem is that both require significant investments in time and money. If you are a small or medium-sized company, you might not be able to afford that, and this is where GenAI presents a unique opportunity.

ERP vs. ML vs. GenAI

• ERPs: These systems require months of studies, blueprints, and implementations before they actually go live. You need to do all the mapping and restructuring of your flows, resulting in a rigid system that is difficult to adjust and co-designed at significant costs. Expensive to set up and maintain.

• Traditional Machine Learning Models: These demand extensive data to make reliable predictions, which means considerable effort in data aggregation and cleaning. Then you need to iterate to build their intelligence with the data, hoping you will be set in a few rounds. In addition, they operate as “black boxes,” providing little insight into their decision-making process.

• GenAI: On the other hand, GenAI already possesses “intelligence” and can mimic human reasoning, enabling accurate predictions with only a small amount of historical data. As it mimics, it can also explain what it does and why.

As we continue to explore the potential of GenAI, automated business flows that were previously out of reach for smaller businesses are finally becoming achievable, and at scale. Isn’t it amazing?

We'd love to hear your thoughts! Can you think of other areas where this could apply?

Hugo Matthaey