Lost Money or a Lack of Collections Structure?

At the end of every quarter, part of the portfolio moves into a different category. Receivables that were “past due” become “difficult to collect” and, shortly afterward, are added to the allowance for doubtful accounts. The explanation in the report is usually the same: the customer didn’t pay. But before accepting that loss, there’s another question few managers ask: did anyone actually try to collect?

At many companies, the honest answer is uncomfortable. Some of the value written off as lost never received an effective follow-up, was never offered a settlement, or came with a promise to pay that no one tracked. That isn’t credit risk. It’s a lack of structure. And the difference between the two determines whether the company is dealing with a market problem or one it can solve itself.

Not Every Loss Is a Loss

There are real losses: a customer that shut down, a receivable that was legitimately disputed, fraud, or a debt that has passed any practical chance of recovery. This is part of any business that sells on credit and should be addressed through credit policies, guarantees, and risk analysis.

Then there are losses created by the process. They happen when a customer could have paid, or negotiated, but the company didn’t reach them at the right time, with the right approach and a viable alternative. This portion doesn’t appear separately on the balance sheet. It gets lumped in with real losses and treated as an unavoidable cost of doing business.

This has a dangerous effect. When a company doesn’t distinguish between the two, it tends to respond in the wrong way. It tightens credit approvals, loses good sales, and continues leaving money on the table because of operational failures that no credit policy can fix.

Five Leaks in an Unstructured Collections Process

A lack of structure rarely shows up as one major mistake. It manifests in small, repeated failures that quietly drain cash when they add up.

Receivables that never get followed up on. In large portfolios, a significant portion of past-due accounts simply never receives a follow-up. The team prioritizes the largest balances or best-known customers, while the rest age without any action until recovery becomes unlikely.

Late follow-up. The likelihood of recovery drops as the payment becomes more overdue. A customer contacted on the third day still has the debt fresh in mind and usually has more options. The same customer contacted on the sixtieth day has probably already prioritized other creditors.

Collections without an offer. Reminding a customer that they owe money isn’t negotiating. When the outreach is limited to notifying them that payment is late, a customer who can’t pay the full amount has nowhere to turn. Without a down payment, deadline, or installment plan within established guidelines, the conversation ends without an agreement.

Promises without follow-up. The customer says they’ll pay on Friday. No one records it or checks on Monday. The promise turns into silence, and the receivable goes back into the queue as if nothing happened—only now it’s a few days older.

Disconnected data. Outdated contact information, a payment that wasn’t posted, a duplicate receivable. Every data-entry error means a collection attempt that doesn’t reach the customer—or, worse, an improper collection attempt that damages the relationship with someone who has already paid.

None of these leaks has anything to do with the customer’s ability to pay. They all have to do with how collections are organized.

How to Find Out How Much of Your Loss Is Structural

Before investing in any tool, it’s worth conducting a simple diagnostic of your own portfolio. The goal is to answer, with numbers, whether the money was lost because of risk or process.

Set aside receivables written off as losses in recent months. Use a six- to twelve-month period and include the amount, due date, and write-off date.
Gather the action history for each receivable. Check how many follow-ups were made, through which channels, on what day of delinquency the first attempt took place, and whether a settlement offer was made.
Sort each receivable into one of two columns. On one side, put cases with documented collection efforts and a clear reason for nonpayment. On the other, put cases with little or no follow-up, no offer, or an abandoned promise to pay.
Calculate the share represented by the second column. The sum of these amounts estimates what the company lost because of a lack of structure. Even if only some of it could have been recovered, this figure is usually enough to change the conversation in the boardroom.

If most losses fall into the first column, the focus should be on credit policy. If a significant portion falls into the second, the problem is in operations—and operations can be fixed.

What a Collections Structure Looks Like in Practice

A collections structure isn’t just having someone responsible for accounts receivable or a well-formatted spreadsheet. It’s a system that ensures every past-due receivable gets the right action at the right time, without relying on someone’s memory or availability.

This system rests on a few pillars: reliable, up-to-date data from the ERP or financial system; a clear policy for discounts, installment plans, and approval authority; a contact cadence based on delinquency stage and customer profile; documentation for every agreement, including records and installment tracking; and real-time visibility into what’s happening across the portfolio.

When one of these pillars is missing, the others lose strength. A well-designed negotiation policy is useless if no one reaches out. A rigorous cadence loses value if payments aren’t identified and customers who have paid in full keep receiving collection notices.

Why Hiring More People Doesn’t Solve the Problem

The natural response to this diagnosis is to expand the team. The problem is that people can handle more volume up to a point, but they don’t solve the consistency issue. Each analyst negotiates differently, works during business hours, and has to split their attention between recording, verifying, and collecting payments. As the portfolio grows, costs grow with it and the leaks continue—just on a larger scale.

Where AI Fits Into the Structure

Cobrança IA acts as the layer that turns structure into continuous execution. Autonomous agents connected to company systems track every receivable starting before it’s due, reach out via WhatsApp, present terms allowed by the commercial policy, document agreements, and track each installment through settlement.

This directly addresses the five leaks. No receivable goes without follow-up because the operation doesn’t depend on manual prioritization. The first contact happens when the collections schedule says it should, not when someone has time. Every eligible conversation can end with a concrete offer. Promises are automatically recorded and monitored. And integration with the financial system reduces improper collection attempts and outdated contact information.

The company stays in control. It sets discount limits, the number of installments, cases requiring human approval, and situations that should be escalated. AI operates within these rules 24/7 and records every interaction for auditing. The finance team stops chasing payments and starts managing exceptions and adjusting strategy.

Metrics That Show Whether Your Structure Is Working

Once collections are structured, the initial diagnostic becomes part of ongoing management. A few metrics help track whether the leaks are actually being closed.

Track the percentage of past-due accounts with no follow-up; it should approach zero. Measure the average time between the due date and the first effective contact. Monitor the rate of offers made per contact and the rate of promises kept. Most importantly, compare the number of receivables written off without a documented attempt to negotiate before and after the change.

This last metric is the most revealing. When it drops, the allowance for doubtful accounts stops concealing process failures and begins reflecting only the portfolio’s actual risk.

The Question That Changes How You Manage Accounts Receivable

Accepting losses is part of selling on credit. Accepting losses that could have been avoided is a choice, even when no one makes it consciously. Before writing off a receivable, the company needs to be able to answer whether it was collected on with structure, context, and a settlement option.

When the answer is yes, the loss is legitimate. When it’s no, the money wasn’t lost. It was left behind. And that’s a problem that can be solved.