Thanks to AI, businesses will get a god-like view into how they spend money

January 4, 2017 Tradeshift Editorial Team

Future of payments

Financial transactions are leaving digital trails for machine learning and AI to roam, which means some day soon finance teams will be left with few mysteries as to where the money goes after it leaves the company coffers.

Every day, employees pay for work needs, such as flights, car share rides, office supplies, and birthday cakes for the boss. And they’re responsible for tracking, approvals, and reimbursements. But there are logical limits when it comes to the details. When was the last time you had to include a receipt for an expensed item under $20? It’s simply unwieldy to manually track and capture the specifics of all company spend because it’s a headache.

Here’s the real question: Does anybody like to drag a set of processes, approval flows, and policies with them as they go about their hectic day focused on what they were hired to do? Not really.

The latest generation of solutions for expense and spend management have helped to peer into “tail” or non-contracted spend, but it seems as if there’ll never be a silver bullet for 100% visibility. Businesses have long ago thrown in the towel. The returns are not understood because 100% has never been achieved. That can soon change.

How Artificial Intelligence can help

AI and its subparts, such as machine learning (ML), deep learning, and data mining are poised to usher in a renaissance in spend visibility. They’ll help you make better decisions in sourcing, budgeting, and approvals, and more. Total spend visibility can only help when you’re facing changes to your business model, need to adapt to shifts in demand and operations, or are under pressure to uncover hidden pockets of potential savings across your business.

Spend visibility on this level requires a comprehensive view into where your company spends all of its money. That includes tracking everything, such as corporate credit card transactions, ACH payments, paper invoices, purchase orders, checks made out to vendors, p-cards, package slips and any other transaction info between you and your suppliers. Whatever is not digitalized needs to be before it’s combined with digital-native data like online payments and electronic invoices.

The heavy-duty task of capturing all financial transactions recorded on paper, emails, PDFs and other unstructured formats is perfectly suited for the algorithmic world of AI and ML. It would take ages for humans to make sense of the volumes of data pouring in from all these sources (and the data is often incomplete) and bring it into a consistent form.

Much like how Google used ML to digitize massive libraries of images and other unstructured assets, companies will be able to use ML to capture the images of the physical and unstructured documents mentioned above, extract the data, and move it all under one roof for company-wide spend intelligence. Once extracted and translated into the language of machines, transaction data can flow into ML algorithms before landing in big data tools, cloud-based data warehouses, and other affordances made possible by today’s immense computing power.

Just imagine if you’d have ALL your spend data extracted, combined, validated, classified, and enhanced with related business information, all with ML automating throughout the process. Your company’s finance and procurement experts would only need to spot-check for accuracy and answer occasional questions to ensure the model kept working. Common errors, costly processing time, and inaccuracies caused by human biases would cease to exist.

Spend classification and other use cases

Take spend classification, for instance, which is one area companies — even those with sophisticated procurement systems — have a hard time understanding. Machine learning algorithms can combine and crunch through all your data sources, including ERP, eProcurement, p-cards, and other external systems and do things like accurately tag line-level details from invoices, ensuring accurate spend categorization.

ML can also combine data from the incoming side of the transaction to the outgoing side to give you insight into the relationships your organization has with your suppliers, customers, and partners. And you’ll be able to use that insight to change how your organization spends to meet your business objectives.

Compared to commonly used approaches today, applying new AI technology to spend visibility would result in the greatest attainable clarity into your cash outflows, spend categories, subcategories, and vendors. Going further, you’ll be able to use that insight to help determine how you build your products and even what to build. You’ll even be able to use ML to help you make decisions based on the insights that ML previously generated!

Thanks to the latest in AI, machine learning, and digital commerce, we are on the verge of a paradigm shift where all spend will be accounted for, bringing with it connected processes and incredible efficiency.


About the Author

Tradeshift connects buyers, suppliers, and all their processes in one global network. We help you transform the way you work with suppliers today – and adapt to whatever the future brings.

Follow on Linkedin More Content by Tradeshift Editorial Team