The Numbers Don’t Lie: AI Maturity in Accounting Demands Leadership

I started my career the way most accountants do. Head down, month-end always looming, not enough hours in the day…or week(s). Reconciliations, journal entries, chasing down data, running the same reports 12 times because something was always off in one of them.

That experience is why we built FloQast. And after a decade in this profession's corner, we’re deeply invested in AI. Every day, I think to myself “I wish this was around when I was a staff accountant.”

But as AI permeates its way through, well, every industry, we wanted to understand how accountants perceive AI.

So we commissioned independent research across accounting and finance professionals in the US and UK, fielded by a third party, with no FloQast filter on the answers.

Here's what they told us.

Six in ten accountants are still spending 40% or more of their time on manual work.

Reconciliations. Journal entries. Transaction matching. Variance investigation. 

Work that is, by definition, automatable. Nearly one in five give up more than 60% of their week to it. Three full days. In 2026.

There's a version of this where 40% manual work today is progress. When I was closing the books, it felt like 100% of my job was manual. So maybe the right reaction is: things are getting better.

But I keep coming back to a more uncomfortable read. We are deep into what everyone keeps calling the AI revolution. The tools exist. The investment is flowing. And more than half of accounting professionals are still giving two full days a week to work that doesn't require a human to do it.

Which brings me to one of the most fascinating numbers from the findings. 

85% of organizations call AI a strategic priority. Only 10% are using it extensively.

I've thought a lot about why that gap exists, and I think the framing is actually part of the problem. "AI strategy" is a phrase that's impossible to act on. 

What does it mean to have an AI strategy? You buy some tools? You run some pilots? You add a competency to your annual review?

None of that produces automation. And automation is the goal. 

The teams seeing real results made a different kind of commitment. They mapped specific workflows to automate. They built the governance before they scaled. They embedded AI directly into the close instead of running it alongside the process. And they did it in that order.

But here's what keeps the accountant in me up at night.

51% of organizations admit their financial controls are informal or inconsistently applied through the close.

This isn't a problem AI created. It's a problem AI is about to make more visible.

The technology is causing a careful profession to move fast, but the worst thing you can do is layer AI on a foundation that was never built to support it. In those cases AI won’t fix a process, it will accelerate your problems. 

When AI gets something wrong in a consumer product, you fix it. When AI gets something wrong in a financial close, the implications can be significant. The profession is right to move carefully.

What isn't sustainable is moving carefully as a permanent state.

Because the data also shows what it looks like when organizations get this right…

The most mature teams in our research close in 6.7 days.

Spending just 34% on manual tasks. 

The least mature close in 8.7 days and spend 63% of their time on manual work. Two full days of close cycle time. Nearly 30 percentage points of manual burden. 

Imagine getting critical financial data to your leadership team two days earlier every single month.

Again, the controls environment matters more than most conversations about AI acknowledge. The teams moving up the maturity curve built governance and documentation first. They started with bounded, well-understood processes. They used their close checklist as a roadmap for what to automate, and they worked through it methodically. The first win is rarely impressive on a slide. But it builds the kind of confidence that compounds.

One final number I’ll leave for the accountants reading this, wanting to get a leg up in the next decade. 

88% of practitioners told us AI literacy will matter as much as GAAP expertise for career growth.

I believe this in the sense that the combination of domain knowledge and AI fluency is where the real advantage lives. But I want to be clear about which one is foundational. You need to know how GAAP applies to your specific business before you can explain it to AI well enough for AI to be useful. The expertise comes first. The fluency amplifies it.

The organizations getting this right have one thing in common. It isn't the technology.

When we looked at what separates the most mature organizations from the ones still stuck, the answer wasn't the tools they bought or the budget they committed. It was the decisions their leaders made before they scaled anything.

One stat stands out to me above the rest: 69% of the most mature organizations have a fully executed AI roadmap. Most of the profession is still figuring out where to start. That divide isn't access to technology, it's leadership.

These numbers are just scratching the surface.

What I find encouraging is that this research doesn't just show us where the profession stands. It gives us a clearer picture of what’s working. Some accounting organizations are already showing what’s possible, and the path they're taking isn't out of reach for everyone else.

We’ll run this research every year because a single snapshot tells you where the profession stands. A trend line will tell us where it’s going. I expect the gap between AI ambition and execution to look very different a year from now.

Until then, we finally have a clear benchmark for where the profession stands today, and evidence of what the organizations leading the way are doing differently. 

Read the full State of Accounting AI 2026 report here.

‍