Author: Marcel Syriani, COO, DATABASICS, Inc.
Expense management problems rarely start where finance teams go looking for them. A few months ago, my team worked with a finance director at a nonprofit. She kept hitting the same policy headache, where employees submitted expenses that should never have been approved. Her proposed fix was to start penalizing approvers whenever a non-compliant expense slipped through.
Punishing the approver after the fact solves nothing. By the time an expense lands on someone’s desk, the money has usually left the account. The employee has already taken the flight, booked the hotel, or run the card. You end up with documentation theater, a review process that looks like control without producing any.
For years, organizations have tried to fix this by piling on rules, approval layers, and complexity. The assumption is simple, that tighter controls produce better financial discipline. In practice, the opposite tends to happen, because most teams are training employees to survive a broken process instead of building one that makes the wrong move impossible.
I call this the Approval Illusion, the belief that more oversight automatically creates more control. In reality, it manufactures delays, workarounds, and hidden costs that never surface on a report.
After decades of building and implementing enterprise systems, I keep seeing the same five mistakes repeat. First, companies design the whole process around the 5% of edge cases, then force the other 95% of employees through needless complexity. Second, finance asks for more visibility, yet employees end up spending more time submitting, correcting, and resubmitting. Third, IT bolts on integrations to improve data flow, and each one quietly adds maintenance cost and new failure points. Fourth, policies grow more detailed while adoption falls and support tickets climb, and when adoption drops, control does not improve, it erodes. Fifth, costs rise across the board, not only in spend but in time, effort, and rework.
Expense Management Breaks Down Because It Reacts Too Late
Expense management breaks down for a structural reason, since traditional systems are reactive by design. They review and correct transactions after the money has already left. By that point, finance is no longer shaping behavior. It is documenting it.
That creates a familiar loop of waste. Teams spend hours auditing rather than guiding decisions. Errors surface late, so rework and corrections pile up. Meanwhile, employees lose productivity fighting through complex steps, and finance scales headcount instead of fixing the system. The result looks controlled on paper, yet it stays expensive to run. That is expense management as cleanup, not control.
Industry reporting points the same way. Recent coverage of automated expense controls shows that fragmented workflows and weak visibility are where most of the risk hides, long before any breach reaches a report.
How Fintech Changes the Model
Fintech did not fix expense management by digitizing receipts or polishing the interface. Instead, it changed the model by moving control upstream while cutting operational cost at the same time.
Modern tools embed policy at the point of spend rather than after it. Rather than leaning on approvals and audits, they shape behavior in real time. Picture how that plays out in practice. A card carries merchant category restrictions baked in at the network level. An employee whose role never requires airfare simply cannot book a flight on that card, because the system declines the transaction before it happens. There is no submission to review, no approver to penalize, no violation to document. The control is structural, not supervisory.
This shift mirrors the broader move toward embedded finance, where financial logic lives inside the tools people already use. Across the wider B2B payments landscape, the same pattern is playing out, as reconciliation and controls get pushed into core business platforms rather than bolted on afterward.
Pair that with real-time integration across card activity, approvals, and expense data, and you get something legacy systems cannot match. Visibility arrives before the damage, not after. A manager sees a spend anomaly the moment it occurs. Policy exceptions surface immediately instead of hiding until next month’s audit. These are the kind of fintech fixes now reshaping modern expense management for teams tired of cleaning up after the fact.
When the system makes the right action easier than the wrong one, you cut manual reviews, remove rework, and shrink the administrative footprint that reactive processes demand. So the gain is not only operational. It is cheaper.
Where AI Fits and Where It Does Not
Here is the counterintuitive part about AI in expense management. It is at once the most promising development and the easiest thing to get wrong.
AI will add real value to expense management. It can spot spending patterns, flag anomalies, suggest categorizations, and lift the manual burden off finance teams. Give it a well-designed process, and it accelerates everything. The technology trims the remaining manual touchpoints, catches the exceptions your rules missed, and gets sharper over time.
But point AI at a tangled process, one with too many validation steps, too many approval layers, and policies nobody can follow, and it simply learns the mess. It will optimize a broken system rather than repair it. Worse, it does that at scale, which means a bad process gets expensive faster.
That tension shows up across the industry. Even as AI in fintech draws enormous attention, the payoff still depends on the process underneath. Finance leaders are starting to treat AI and automation as a way to move from reactive to predictive operations, though only when the foundation is clean.
So the real opportunity for expense management is to pair AI with better controls and smarter automation. Use it to reduce unnecessary validations instead of adding more checks. Simplify policies so the system can enforce them automatically. Strip out manual touchpoints that add no decision value, then design processes that lower effort before anyone tries to optimize them. AI should cut costs through clarity, not paper over inefficiency.
The Real Cost of Expense Management Is Not What You Think
Finance leaders tend to measure expense management by what gets spent. That is the wrong metric. The real cost sits in the operational overhead of running the system itself, the hours your team loses chasing receipts, the productivity employees burn navigating complexity, the rework when errors surface too late, and the headcount you add to babysit a process that a better design would never need.
It is the same blind spot that shows up elsewhere in finance, from cash flow to the hidden drag of a late payment crisis, where the visible number quietly hides the expensive one.
When expense management shifts from a cleanup function to a decision function, the impact reaches well beyond visibility. Teams stop chasing transactions, overhead drops, and unnecessary work disappears across the organization. Rather than layering on more controls, you build systems people can follow.
So consider that nonprofit director again. The answer was never to punish approvers. It was to ask a sharper question, namely why non-compliant expenses were reaching the approval stage at all. Fix that, and the approver problem dissolves. The control does not get stricter. It gets earlier.
That is where real cost control lives, not in tougher policies but in better-designed processes. Because the most expensive system is never the one with the highest spend. It is the one that takes the most effort to manage.
Marcel Syriani serves as Chief Technology Officer and Chief Operating Officer at DATABASICS, where he oversees operations, finance, product direction, and IT. DATABASICS delivers intuitive, powerful solutions that take the hassle out of time tracking and expense reporting, helping organizations streamline operations, ensure compliance, and support teams wherever they work.
