It starts with a payslip. A sales representative opens the monthly commission statement, looks at the number and feels something drop. It is not what they expected. They check their own notes, count the deals they closed, and arrive at a figure several thousand rand higher. They raise it with the sales manager, who raises it with finance, who opens the spreadsheet and sees that the numbers match the source data. The source data, it turns out, is wrong. Two deals were booked under the wrong person, a customer was recorded twice, and one sale was dated the day after the cut-off.
This piece looks at why commission disputes so often originate in the data rather than the people, what that does to trust, and how to make the calculation something staff can follow and believe. It is a hypothesis about a common pattern, not a claim about any particular company.
Why the data goes wrong before the calculation starts
A commission payment is the end of a long chain: a deal is agreed, recorded in a sales system, matched to an invoice, confirmed as paid, assigned to a person and a period, run through a rate table and passed to payroll. A fault at any link carries through to the payment, and the person who sees the result is the last to find out.
The common faults are mundane.
- Ownership recorded wrongly. Accounts change hands, deals are shared, and the customer record still carries the previous owner.
- Dates that disagree. The order date, invoice date and payment date differ, and the commission plan does not say which one counts.
- Duplicates and returns. A repeated entry inflates one person's figure; a credit note is never linked back to the original sale.
- Rules living in a spreadsheet. Rates, caps, accelerators and exceptions are encoded in formulas that only one finance person understands.
Mellorca's analysis of fees, commissions and charges calculated in spreadsheets covers the error risk in more detail, and sales and finance using different product and price data shows how the underlying disagreement arises.
What the evidence can and cannot say
No South African study of commission disputes is cited here, and no error rate is claimed. The argument rests on general principles.
Reference material on sales commission describes it as a form of payment linked to sales activity, typically a percentage of the value of a sale, and notes that the design of the scheme shapes behaviour. When pay depends on figures, the integrity of the figures becomes inseparable from the fairness of the pay.
General references on data quality describe it as the fitness of data for its intended use, assessed along dimensions such as accuracy, completeness, consistency and timeliness. A commission run needs all four, and it fails visibly when any is missing. The older principle often summed up as "garbage in, garbage out" makes the same point: a flawless calculation applied to flawed inputs still produces a flawed result.
In South Africa, the Basic Conditions of Employment Act sets rules on how remuneration is paid and on what an employer may deduct from it. Recovering an overpayment, for example, is not simply a matter of reducing next month's payslip. That adds a legal reason to get the data right the first time, and it is a reason to take advice from an employment specialist when a correction is needed.
An illustrative case
Consider an illustrative example, not a real case: a Johannesburg office-furniture supplier with eight sales representatives and a two-person finance team. Commission is a percentage of the invoice value, paid in the month after the customer pays. The commission statement is built each month by one finance clerk from an export of the sales system and a list of payments from the bank.
In one month, a representative named Thandi sees a statement that is R6 400 lower than her own count. She asks the sales manager. The sales manager asks finance. Finance rechecks the formulas, finds nothing wrong and says so. Thandi, who is told that the calculation is correct, concludes that she is being underpaid on purpose or that nobody believes her.
Two weeks later, a finance clerk discovers that a large order was recorded against a colleague who had previously managed the account. The calculation was right. The input was wrong. By then Thandi has told two colleagues she is thinking of leaving, the sales manager has stopped trusting the monthly report, and a dispute that could have been resolved in an hour has taken most of a fortnight.
The human tax: confidence and financial resilience
The first cost is confidence. People who cannot check how their pay was worked out must choose between trusting a figure they cannot verify and doubting colleagues they work with every day. Either choice damages the relationship. The Mellorca article on when nobody can say which number is true describes the wider effect of competing versions of the facts.
The second is financial resilience. Commission is often a large share of a representative's income, and households plan around it: a bond payment, school fees, a vehicle instalment. A shortfall of a few thousand rand, even if later corrected, can cause real strain in the meantime. A business that treats a correction as a routine adjustment may not see how heavy the delay is for the person waiting. When late invoices become late salaries covers a related effect from the employer's side.
What could realistically change
The aim is a commission process in which any person can see how their figure was built and any discrepancy can be traced to its cause.
- Write the plan in plain language. State which date triggers the commission, what happens with returns, how shared deals are split, and who decides edge cases.
- Fix ownership at the source. Make the person responsible for each deal a required field in the sales system, and review ownership changes at month end rather than after payroll.
- Share a pre-payment statement. Send each person a line-by-line breakdown a few days before payroll closes, so errors are found while they are cheap to fix.
- Reconcile before calculating. Check totals between the sales system, the invoices and the payments list before the rate table is applied. The article on manual reconciliation between systems explains why that step matters.
- Keep a dispute log. Record each query, its cause and its fix. After a quarter, the pattern shows where the data most often goes wrong.
The business outcome
Disputes are shorter and rarer, finance spends less time on investigation, and managers can rely on the commission report as a record of what happened. Over time, the sales team's attention moves from checking the payslip to selling.
The human dividend
The dividend is predictability and breathing room: knowing that a payment will be right, or that if it is wrong, there is a clear way to find out why. For a household that budgets around variable pay, that reliability is worth more than a modest rise in rates.
The trade-offs
- Transparency exposes disagreements. A clearly written plan will show where people interpret rules differently, and some of those conversations will be difficult.
- Pre-payment statements add a step. They require finance to finish earlier, which may compress an already tight close.
- Tighter data rules slow sales admin. Required fields can irritate representatives who want to move quickly.
- Not every dispute is a data fault. Some are genuine disagreements about the plan, and those need a decision, not a correction.
Why it matters
When a commission dispute starts in bad data, the person who pays for it is the one who trusted the system. Making inputs correct and the calculation visible protects both the accuracy of the payment and the relationship it is supposed to reward.
Sources and further reading
- Commission (remuneration) - Wikipedia (last verified 10 October 2026)
- Data quality - Wikipedia (last verified 10 October 2026)
- Garbage in, garbage out - Wikipedia (last verified 10 October 2026)
- Basic Conditions of Employment Act 75 of 1997 - South African Government (last verified 10 October 2026)
Related Mellorca reading
- Why Rest Days Start With a Status Check
- The Employee Who Needs Permission for Every Small Fix
- The Founder Who Wants to Step Back but Can't