New research reveals where finance teams see the biggest opportunities for AI in accounts receivable and what’s holding them back.
AI has made its way into almost every finance conversation.
But knowing that AI can be used in finance is very different from knowing where it can actually make a difference.
That’s the more useful conversation for accounts receivable teams.
Finance teams already have technology supporting their day-to-day work, from the ERP at the center of financial operations to the systems and tools that support accounts receivable. The opportunity with AI isn't simply adding more technology. It's using AI where it can take meaningful work off the finance team's plate.
The real value comes from knowing which work makes sense to hand over to AI, which work it can help with, and where people still need to make the call.
New research from Flywire, a Founding Sponsor of The ERP Update, gives us a closer look at where finance professionals see that opportunity.
Flywire surveyed 300 U.S. finance professionals responsible for accounts receivable, cash flow, and collections. Two-thirds are actively exploring AI solutions, and 90% have a budget allocated. At the same time, one-third say the ROI still isn’t clear, and they need guidance on where to begin.
That combination tells an interesting story.
Finance is ready to invest in AI. Now it needs to decide where AI can earn its place.
Before thinking about AI, start with the work.
Despite years of investment in ERP and automation, familiar manual bottlenecks remain in A/R.
Flywire’s research found that data entry between systems and following up on overdue invoices were each cited by 26% of respondents as significant manual bottlenecks. Cash application and reconciliation followed closely at 25%.
These are exactly the kinds of areas worth examining first.
The goal isn’t to add AI everywhere. It’s to identify repetitive work that continues to consume the finance team’s time and determine whether AI can take some of that work off their plates.
One of the most interesting findings in Flywire’s research is how broadly finance professionals are thinking about AI.
They aren’t only interested in using it to generate reports or summarize information.
Among the capabilities respondents expressed interest in:
There’s a common thread running through that list.
Finance teams want help with the work that happens between receiving information and deciding what to do with it.
Matching a payment. Prioritizing an account. Pulling information from a document. Drafting a routine message. Understanding what deserves attention first.
That’s a much more interesting use of AI than simply giving finance teams another place to type a prompt.
Finance teams have been automating processes for years. AI introduces something different.
Traditional automation generally follows predefined rules: when X happens, do Y.
Flywire’s report describes AI as moving beyond those rules by using context, identifying patterns, reasoning about what’s happening, and potentially recommending or taking the next action.
Collections offers a simple example.
An automated workflow might send a reminder seven days after an invoice becomes overdue.
AI could use available payment history and account information to help determine which overdue accounts need attention first or assist with creating the appropriate communication.
Automation completes a predetermined task.
AI can help determine what should happen next.
That distinction is where some of the more interesting opportunities for finance begin.
Giving AI a larger role in financial processes also changes the conversation around trust.
Flywire found that trust and transparency were the top concern about AI adoption, cited by 39% of respondents. Accuracy followed at 37%, integration with existing systems at 35%, data privacy at 34%, and limited internal expertise at 31%.
Those concerns become especially important as AI moves from helping complete routine tasks to recommending or taking financial actions.
Not every decision should be handed over completely.
Flywire’s research points to human oversight, clear guardrails, audit trails, and explainable actions as important considerations for higher-stakes decisions.
This is where one question is worth keeping: Can AI do this and should AI do this on its own?
The answers may be very different.
For finance teams trying to decide where to begin, the best starting point may not be an AI product.
It’s the current A/R process.
Look for the repetitive tasks that happen hundreds of times each month, information that has to be moved manually, and work that routinely gets pushed aside because the team simply doesn’t have enough time.
Flywire recommends beginning with high-volume, lower-risk workflows such as payment matching, reminder sequences, and data extraction. These areas give organizations an opportunity to prove value before applying AI to more complex or higher-stakes decisions.
There’s another consideration that matters for ERP teams: AI shouldn’t become one more disconnected system.
Flywire recommends looking for AI capabilities that integrate with an organization’s A/R platform and ERP rather than simply adding another standalone tool to the finance technology stack.
The ERP continues to provide the financial foundation. AI can help finance teams do more with the information and workflows around it.
With 90% of respondents reporting that they have budget allocated to AI initiatives, investment is clearly moving forward.
But the first conversation probably shouldn’t be: “Which AI solution should we buy?”
It should be: “What job do we need AI to do?”
Maybe it’s matching payments.
Maybe it’s helping prioritize collections.
Maybe it’s extracting information from documents, assisting with forecasting, drafting routine communications, or helping the A/R team find answers faster.
Those are specific jobs. And specific jobs can be evaluated.
Finance teams can measure how much time the work takes today, determine what level of human oversight is appropriate, and see whether AI actually improves the process.
Because finance doesn’t need AI everywhere.
It needs AI where it can earn its place.
This article is based on research from Flywire's Beyond Automation: What Finance Teams Really Need from AI, which surveyed 300 U.S. finance professionals responsible for accounts receivable, cash flow, and collections. Flywire is a Founding Sponsor of The ERP Update.