Shadow accounting is the practice of salespeople keeping their own record of what they believe they are owed, usually in a private spreadsheet, alongside the employer's official commission system. It is a direct measure of how far the official statement is trusted, and the hours it consumes come straight out of selling time.
The file is called something like My Numbers or commissions_2026_v4, it lives in a personal Drive folder, and it has a tab per quarter going back further than the rep has been at the company. Every deal is in there with the date it closed, the rate the rep believes applies, and a running total. On payday the rep compares their total to the statement, and the gap decides whether the afternoon is spent selling or emailing.
Nobody builds this for fun. A rep whose variable pay is forty percent of their income is carrying a month's mortgage on somebody else's arithmetic, and they have almost certainly been shorted at least once. Auditing is what people do when the cost of being wrong is high and the ability to check is low. The spreadsheet is a rational response to a statement that shows a total with no workings behind it, and the reps who keep the most detailed ones are frequently the strongest performers, because they have the most at stake.
The second-order effect is the expensive one. Once the rep's sheet and the company's statement disagree, the argument is between two sets of books, and the rep's assumptions are unwritten. They may have used the accelerator rate on a deal that fell below the threshold, or counted a booking the plan pays on invoicing. Neither side can adjudicate, so it resolves by seniority rather than by evidence, and both parties leave more suspicious than they arrived. A statement that shows the deals, the rule applied to each and the arithmetic makes the private copy redundant. Every other approach leaves the rep with a reason to keep it.
A sales director asks around and finds 22 of 30 reps keep their own commission tracker, each spending roughly three hours a month maintaining it and checking it against the statement. The costed hourly rate here is an assumption; substitute your own.
That figure counts only the reps' own time. It excludes the operations hours spent answering the queries those spreadsheets generate, and it excludes whatever the same 792 hours would have produced if they had been spent in front of customers.
The instinct, on discovering the trackers, is to tell people to stop and trust the system. It never works, and it costs something: the spreadsheets go underground, so the mismatches that used to surface as questions now surface as resignations. Shadow accounting is a measurement, and the useful response is to read it. Ask three reps to show you their file and mark where it diverges from the statement. Those divergence points are a precise list of what your commission process fails to explain, gathered by the people with the strongest possible incentive to be thorough.
Commish pays shadow accounting the way your plan describes it, shows the arithmetic on every line, and traces each payment back to the deal that earned it.