“A quicker dashboard doesn’t make you strategic. A better call does.”
Rita Maria Reed has spent more than 20 years in finance, including leadership roles at Microsoft and Accenture. And if there’s one thing she’s learned about transformation, it’s that adding new technology doesn’t automatically change how work gets done.
Rillion’s 2026 AI in Finance Report shows that 68% of US finance leaders expect AI to significantly or fundamentally change the finance function within three years. Rita believes getting there means going beyond small pilots and looking at whole areas of work: what can the system take on, where do people add the most value, and what can you stop doing altogether?
What does a truly scalable finance function look like in an AI era, and how far away are most organizations from that reality?
Scaling finance used to mean adding more people and more duct tape. More headcount, more spreadsheets, more patches holding it together. In the AI era, it’s the opposite. The work grows faster than the headcount. Routine steps are handled by the system, exceptions go to people, and humans focus on what requires real judgment.
How far away is that? Starting is closer than most think. Scaling is farther than the hype suggests. You don’t need perfect data to begin. You need data you can check, and more finance leaders have that than they realize.
68% of US finance leaders expect significant or transformational change within three years, but among organizations not using AI, that drops to 42%. Why does experience change expectations so dramatically?
Because using AI changes what you see. Before you try it, AI can look like a tool for writing things faster. Once you apply it to real work, you start seeing the whole process it could change: the steps, the checks, the decisions.
The leaders using AI more widely have seen what happens when it moves beyond writing and into real work. Others are still trying to picture it. Once you’ve seen what’s possible, change stops feeling optional because now you can also see the cost of leaving the work exactly as it is.
When it comes to AI, what is the single biggest mistake finance leaders make when trying to scale from pilot to transformation?
Chasing scattered little wins instead of rebuilding one whole area of finance. You spread AI across a dozen small tasks, keep people in the middle doing the real work, and the pilot stays a pilot because nobody built the controls or ownership to run it for real.
It looks busy, but nothing really changes. Instead, pick an area, decide what it should become, and rebuild it end to end. Let the system handle the routine work and move people up to where judgment matters.
I learned this from a compliance problem. The job was simple: which deals should I review for quote authority. So we built automated dashboards to catch them. They improved discount compliance, then showed me a way to take two days out of approvals. I went in to catch the wrong deals and came out with a plan to move more good quotes through, faster. That was my first aha with domain thinking.
57% of US finance leaders expect AI to give them a stronger role as strategic decision-making partners. Is that actually happening, and what has to change to make it real?
It’s starting to happen, but mostly where finance uses AI to make better decisions, not just faster reports. A quicker dashboard does not make you strategic. A better call does. Finance needs to connect day-to-day activity to the dollars and use consistent definitions so the answer doesn’t change depending on who ran it. AI also has to show its work: the assumptions, the proof, the exceptions. And a person still has to challenge the answer and own the decision.
Trust has to be earned, and hope is not a track record. Once finance can stand behind the number, it stops reporting the business and starts steering it. That’s not a seat AI hands you. It’s one finance earns.
How does the transformation journey look different inside a tech-native company like Microsoft or Accenture compared to a more traditional finance function?
The difference isn’t that tech people are braver. At Accenture, getting people off Excel and onto SAP twenty years ago was a fight. At Microsoft, it took a snowstorm to finally move us from Skype to Teams. Even at the frontier, people cling to what’s comfortable.
What’s different is where they start. Tech companies have speed and engineering resources, so the business pulls finance along with it, but they still need to build the right controls. Traditional finance teams often have tighter process and clearer control ownership, but older systems and less urgency.
Neither is necessarily ahead. They start from different places. Whichever side you’re on, speed needs monitoring and controls underneath it. Otherwise, it’s just a faster way to be wrong.
Building AI you can trust forces you to write down the rules, exceptions, and judgment you’ve kept in your head for years. That can become the training material we never had. Our job is to turn it into a new way of growing people.
What is the conversation you keep having that nobody is writing about yet?
Finance leaders have to grow people differently, and most of us were never taught how. For years, judgment was built almost by accident. You gave a junior the repetitive work, and through that exposure they learned how the numbers behave and when something feels off. That grunt work is exactly what AI is starting to take away.
Now we have to teach judgment on purpose. The upside is that building AI you can trust forces you to write down the rules, exceptions, and judgment you’ve kept in your head for years. That can become the training material we never had. Our job is to turn it into a new way of growing people.
If you could give one piece of advice to a CFO who has the ambition for transformation but keeps getting pulled back into the day-to-day, what would it be?
You won’t beat the day-to-day by finding more hours. You beat it by asking the question nobody wants to ask: what are we going to stop doing? Every finance team carries work that has outlived its purpose. Hold a regular stop-doing review, make it someone’s job to retire that work, then protect the time you free instead of letting it fill back up.
Use that time to pick one area where the rules are clear, the pain is real, and you can check the output, then rebuild it properly. You don’t find time for transformation. You fund it with the work you’re willing to stop.
The interview was first published in Rillion's AI in Finance Report.
About Rita Maria Reed
Rita Maria Reed is a global finance executive and leadership speaker with 20+ years of experience across financial services and technology. A former CFO and finance leader at Microsoft and Accenture, she helps organizations lead with clarity, strong judgment, and people-first decision-making in times of change.
