5 key metrics to optimize your accounts payable process

Do you ever feel like you’re spending too much time on specific manual tasks every week or month?

Do you wish you had more visibility into your team’s performance to see why you’re lagging behind?

According to IFOL’s AP Automation Trends 2022 report, 56% of accounts payable teams are spending more than 10 hours per week processing invoices and administering supplier payments. 

In 2019 and 2021, one of the top challenges within AP processes was too much manual data entry. If finance teams are aware they are allocating too much time towards invoice management, why hasn’t this age-old problem been solved yet? The answer to this question is another challenge reported by AP professionals: lack of visibility. 

Whether you have some automation or none, visibility into performance is vital to know where you are, to then use those benchmarks to determine where you want to be.

The importance of accounts payable KPIs

Measuring how much of your process is truly automated might make you rethink your digitalization plan. If there is an opportunity for low risk or recurring invoices to require no human review, tracking how many invoices are processed with no touch will be a game changer and allow you to see true automation progress. 

Paper-based processing and manual data entry is time-consuming and difficult to interpret. When automation comes into play, lack of analytics can make it impossible to understand why specific team members are processing slowly. Is it because they take two hour lunches? Or because they simply don’t understand the system? No wonder these possible scenarios are keeping you up at night. You are unable to pinpoint the specific issue.

In addition to better sleep, leveraging accounts payable KPIs can help you address these common challenges:

• Identify bottlenecks in your process
• Derive valuable information from financial transactions in real-time
• Make informed decisions faster
• Enable sustainable change management
• Decrease invoice processing time

The following KPIs are the starting points to shifting your accounts payable team from a cost center to a profit center. 

(1) # of Autopiloted invoices 

As it is today, does your Accounts Payable team know the number of invoices that are posted in a day without any human review? With explainable artificial intelligence (XAI), autonomy and intelligence to cost accounting processes is possible. AI learns from your supplier’s invoices and can then predict correct values from an invoice to code the data directly into your GL without manual data entry. When the system selects invoices meeting confidence level; it enters the data, classifies the costs, and sends it for approval without human review. This functionality is called what Vic.ai has coined Autopilot, which significantly reduces invoice processing time. 

Consider what differentiates true AI from other popular automation strategies, like robotic process automation (RPA). RPA is rules and template-based, and therefore cannot recognize new invoice fields from the same supplier and adjust to maintain autopilot. Autonomous systems like Vic.ai do not require human input or supervision. This number of Autopiloted, or no review invoices, should then increase over time as the AI continues to learn from your suppliers invoices. 

(2) Processing time per invoice - is your AP team getting faster?

The average time to process an invoice is about 10 minutes with old technologies like EDI, OCR or RPA. This is due to complex approval cycles, lost invoices, manual invoice routing, and manual line level coding. Discounts are also often missed from included early payment programs. Knowing your processing time per invoice is critical to evaluating if your AP team is behind industry standards or ahead. Best-in-class processing time per invoice is 1-2 minutes per bill, which has decreased significantly from days to under a minute as automation tools have become more advanced.

<

Content for Finance Pioneers: Why CFOs are equipped to pave the way for digital transformation

(3) Accuracy rate

When artificial intelligence systems begin to take over the manual tasks of invoice processing, we humans need a way to validate the AI’s confidence level when reading an invoice. Accuracy rate is a way for the AI system to communicate with the accountant. It lets them know it has reviewed all the data and decided it’s confident the invoice does or doesn’t require human review. 

Vic.ai’s explainable AI (XAI) has red, yellow, or green bars that indicate the confidence range of how accurate extracted invoice data points are. Data is only useful if it tells a true story, and to reach an autonomous accounting process, coding accuracy must be prioritized. Unlike RPA or basic automation, corrections can be made to the AI’s initial inaccurate predictions, fine-tuning the dynamic workflow between your accounting organization and our platform.

Minimizing your time spent on invoice processing directly relates to accurate invoice field entry and line item matching. Training is a key component of accuracy as AI engines need data sets to learn from and improve decision making for the next invoice. As the system processes more invoices and algorithms are improved, it generates a positive feedback loop, which increases accuracy and confidence levels over time.

(4) Cost to process a single invoice

Wage rates for labor is a key factor to consider when evaluating the average cost to process a single invoice. Based on labor alone, each invoice can cost several dollars to process. If you process thousands of invoices, this could cost hundreds of thousands of dollars each year. To calculate this statistic, divide your total number of invoices paid by all costs incurred to pay them. This will give you the cost per invoice. This figure should then give a sense of where your AP team is from a cost center or profit center perspective. 

Shifting focus away from low value data entry towards more valuable consultative work can create new revenue opportunities as well as raise your groups operating margins. An AI accountant can streamline the heavy lifting for routine and tedious invoice data entry and posting. An autonomous accounting function not only frees your AP team from monotonous assignments, but statements require less review all together. In summary, the process of paying invoices should not cost more than the invoices themselves.

(5) Volume of invoices to time per invoice ratio

62% of AP teams report they are not well equipped to handle a sudden influx of invoices. A recent IOFM survey found that invoice volumes are increasing significantly as organizations see their business rebound from the pandemic roller coaster. Due to staff downsizing that occurred, there is a huge gap to fill with more invoices to process. 

Though 28% of AP teams are adding staff to handle this volume and 26% are adding automation, almost half of respondents said they are working longer hours to process all the invoices. 

Adding more staff or asking AP staff to work longer hours is not cost effective or sustainable and can likely cause burnout, resulting in turnover. Working longer hours and processing more invoices per day is a recipe for disaster as more coding errors are bound to occur, which will cause late payments or inaccurate financial reporting.

Another advantage of having an AI based accounting system is that as your business grows, your time per invoice will continue to decrease. Hiring more clerks to meet this growing demand is no longer necessary, and can prevent overbearing your staff or suffering from labor shortages. What percent of your accounts payable team is used towards coding invoices? How are they able to keep pace with a growing business?

As you can see in the graph above from a Vic.ai user’s data in 2021, the system can handle hundreds of thousands of invoices per quarter and time per invoice remains under 1 minute. This ratio is easiest to understand with the Vic.ai Analytics dashboard to compare invoice volume and time, rather than computing a specific formula.

Look under the hood of your of your accounts payable operation

Artificial intelligence and machine learning is here and already reshaping many aspects of our professional lives. In accounts payable, efficient teams leverage these technologies and employ key performance indicators (KPIs) to provide a baseline to understand their performance and track operational improvements. Leverage autonomy and analytics to uncover what aspects of your AP workflow delay invoice processing and what should be improved upon. 

Are you ready to enter the new era of accounting?

Profound changes were starting to surface even before the pandemic. Remote work and shifted perceptions of such corporate arrangements has accelerated this digital transformation along with many other business operations. Artificial intelligence continues to play a major role in this direction and is driving new opportunities while raising accounting’s operational margins. 

An autonomous accounts payable function is only possible by harnessing AI, and is the next generation of technology surpassing rules-base automation. The fear of replacement is often unwarranted as AI improves productivity instead of making employees obsolete. In your business, it is vital to consider the functions that AI can do better than humans, and where humans can do better than AI.

A smooth functioning accounts payable team is what distinguishes this department from a cost center to a profit-generating center. The leading enterprises we work with are years ahead of their competitors and have achieved returns on their investment far beyond anything possible with legacy solutions like OCR, EDI, and RPA. 

Begin your autonomous accounting journey with a solution that understands and works alongside your AP function, and is not limited to a predetermined set of rules that in many ways further complicate accounting.

‍