The account that was fine until it was not
A Cape Town building-supplies wholesaler has 340 active trade accounts. In March, one contractor paid ten days late. In April he paid three weeks late. In May he ordered more than ever, and in June his cheque bounced. By then he owed R186,000, three further deliveries were on the road, and the credit controller found out only when the bank returned the payment. Looking back, the warning signs were all there. They were just spread across statements, delivery notes and an accounts clerk's memory, and nobody was positioned to see them together.
This is the pain of not knowing which customers are becoming payment risks. It is different from simply being paid late. Late payment is the event; the pain is the lack of an early view of who is heading that way, so that action happens while it is still cheap.
Chasing invoices consistently is a related but separate subject, covered in inconsistent payment follow-up and its cash-flow cost. The issue here comes one step earlier: seeing which accounts deserve attention at all.
Why risk stays hidden until it is a loss
Most businesses can produce an ageing report that shows what is overdue today. That report is backward-looking. It tells you which invoices have already failed, not which accounts are deteriorating. A customer who used to pay in 30 days and now pays in 45 is still "current" on many ageing reports. A customer whose orders have suddenly doubled while payments slow down is more dangerous than one who is simply late, but no single report shows the combination.
Information that would reveal the pattern lives in different places: payment history in the accounting system, order volumes in the sales system, promises to pay in emails and phone notes, and disputes in a shared inbox. The people who sell to the customer rarely see the payment behaviour, and the people who chase payment rarely see what is being ordered. The story about credit checks in why credit checks happen after sales has promised delivery describes the same disconnect at the point of sale.
Why businesses tolerate it
- Relationships feel like protection. Owners trust customers they have known for years and do not want to appear to doubt them.
- Sales and credit control pull in opposite directions. Nobody wants to be the person who stops a large order.
- The ageing report seems sufficient. It is the report everyone has, so it is assumed to be the report that matters.
- Warning signs are subtle. A few days' slippage looks like noise until it is a trend.
- Review is manual. Looking at every account every month takes time the finance team does not have.
Operational symptoms
- the first time anyone senior hears about a problem account is when a large amount is already overdue;
- credit limits set once at onboarding and never reviewed;
- customers on stop credit still receiving deliveries because the warehouse was not told;
- promises to pay recorded in emails or notebooks but not in the accounts system;
- finance and sales giving different answers about the same customer's position;
- month-end meetings that review the total owed but not individual customer trends.
How the pain becomes money
- Bad debt: amounts that are never recovered, or recovered only after legal costs and discounts.
- Cash-flow strain: money tied up in overdue accounts has to be replaced by overdraft or supplier delay, both of which cost money.
- Added exposure: continued deliveries to a failing account increase the size of the eventual loss.
- Collection effort: staff time spent chasing large, old debts is far greater than the time needed to manage a small, early slippage.
- Tax and accounting effort: writing off debts and correcting records takes finance time and attention, and South African businesses should take advice from their accountant or SARS guidance on VAT on how irrecoverable amounts are treated.
An illustrative cost model
This example is illustrative. Every figure is an assumption for a hypothetical business, not an industry statistic. Picture a trade wholesaler with R40 million annual turnover on credit terms.
| Assumption | Value |
|---|---|
| Annual credit sales | R40,000,000 |
| Average debtor days today | 48 |
| Debtor days if risky accounts were acted on early | 42 |
| Cost of funding (overdraft rate assumed) | 12% a year |
| Bad debts written off per year | R320,000 |
| Share avoidable with earlier action | 40% |
| Credit controller time spent on late, large recoveries | 6 hours a week at R300, 46 weeks |
- Extra funding cost: R40,000,000 ÷ 365 × 6 days × 12% = about R78,900 a year
- Avoidable bad debt: R320,000 × 40% = R128,000
- Collection effort on late recoveries: 6 × R300 × 46 = R82,800
- Illustrative total: about R290,000 a year
The figures are small compared with turnover, but they are comparable to a full salary and fall straight to the bottom line. One large failure can exceed the whole annual total.
Secondary consequences
Hidden payment risk also distorts decisions. A salesperson who does not know a customer is struggling will happily offer a discount to win a bigger order. A buyer who learns late that a supplier is under pressure may be forced to find an alternative in a hurry. Owners who feel burned by one bad account often overcorrect, tightening credit for everyone and damaging good relationships. Where credit is granted to individuals as consumers rather than to companies, the National Credit Act, available from the South African Government documents page, regulates consumer credit and may set rules the business needs to follow. It is worth checking which rules apply to your customers.
When it becomes commercially urgent
- debtor days are rising for several months in a row;
- a single customer represents a large share of the debtors book;
- the business is relying on overdraft to cover timing gaps;
- an industry downturn, a major customer failure or a rise in interest rates puts customers under pressure;
- the business plans to grow through bigger orders on credit.
What better looks like
The aim is a short, regular list of accounts that need a conversation, produced without anyone having to hunt for it.
- Define what "worsening" means. Examples: payment days up 10 days against the customer's own average, a promise to pay broken twice, an order far above the credit limit.
- Combine the data. Bring payment history, orders, limits and notes into one view per customer, even if that begins as a single maintained spreadsheet.
- Produce an early-warning list weekly. Dashboards using tools like Power BI can bring data from several sources together for this.
- Automate the alerts. A workflow, such as the cloud flows described in the Power Automate documentation, can notify the credit controller and the account owner when an account crosses a threshold.
- Link to the order process. Orders for flagged accounts should trigger a review before dispatch, not after.
- Record every promise. Promises to pay belong in the customer record, with date and amount, so broken promises become visible.
The related article on manual payment matching and reconciliation is relevant too, because accurate risk signals depend on payments being applied to the right invoices quickly.
Practical next actions
- List your ten largest debtors and note, for each, how their payment days have changed in the last six months.
- Write a one-paragraph definition of "payment risk" that sales and finance both agree to.
- Check which customers are over their credit limit today and who approved it.
- Agree who may release stock for an account that is flagged, and how that is recorded.
- Build the first weekly early-warning list, even if it takes a clerk an hour to compile.
- Review the result after three months and decide whether automation is justified.
Bottom line
Customers rarely become bad debts overnight; they drift. The business that spots the drift pays for a conversation, while the business that discovers it at the bounced cheque pays for the debt. Building a simple, regular view of which accounts are changing is one of the lowest-cost ways to protect cash flow, and it works best when it connects the people who sell to a customer with the people who collect from them.
Sources and further reading
- National Credit Act 34 of 2005 - South African Government (last verified 10 October 2026)
- Value-Added Tax - South African Revenue Service (last verified 10 October 2026)
- What is Power BI? - Power BI (last verified 10 October 2026)
- Overview of cloud flows - Power Automate (last verified 10 October 2026)
Related Mellorca reading
- Paying the Same Supplier Invoice Twice: The Direct Cost of Weak Controls
- Typing Invoices From Data You Already Hold: The Labour and Billing Delay Cost
- Retyping Invoices Into the Accounting System: The Labour and Error Cost