Trick or True: Unmasking tall tales in your data
The bad data is coming from inside the spreadsheet. Zeros masquerade as no demand, rounded numbers appear in suspicious formation, and operations turn picture-perfect when the clipboard appears. Your dashboard may be more messenger than menace. Before blaming the ghost in the machine, check where the mischief began.
Zeros in Groucho Glasses – Not every zero is as it appears; some are impostors in bargain-bin disguises. A reported zero may mask a station closure, a stockout, a frozen POS, or a missing entry. Treating missing values as zeros can give your analysis the funhouse-mirror treatment.¹ Log those exceptions alongside the POS report. The reason code follows the opportunity to sell, not the ghost of a transaction that never happened. If an item was offered, available, and nobody bought it, that is a true zero. Anything else is an operational exception masquerading as zero demand.
The Truth Is Rounded – Five pounds. Ten minutes. Fifteen ounces. Has your operation achieved eerie precision, or have zeros and fives taken over? When reported values repeatedly end in 0 or 5, response heaping, the human tendency to favor rounded numbers when estimating, may be lurking.² Each month, sort the entries by their final digit. A cluster of zeros and fives is not proof of funny business; it’s a cue to check the equipment, instructions, or whether estimates are replacing measurements. Sometimes the scariest numbers are the ones that look a little too Stepford-perfect.
Somebody’s Watching Me – When the audit clipboard appears, labels straighten, portions become precise, and procedures are executed like magic. That does not mean anyone was hiding a horror show; observation itself may change behavior.³ One formal audit may capture a dress rehearsal, not everyday service. Add five-minute team “friction checks” after selected meal periods to ask, “What made the right way harder today?” Pair answers with operational data, then fix the obstacle, not the employees. The real monster may be missing equipment, poor layout, or lunch-rush time pressure.⁴
Bad data rarely rattle chains; they arrive polished, plausible, and ready for the dashboard. We can help you keep questionable data from haunting your next business decision. Just reach out to us here.
1. Peng, J., Hahn, J., & Huang, K.-W. (2023). Handling missing values in information systems research: A review of methods and assumptions. Information Systems Research, 34(1), 5–26.
2. 1. Holbrook, A. L., Anand, S., Johnson, T. P., Cho, Y. I., Shavitt, S., Chávez, N., & Weiner, S. (2014). Response heaping in interviewer-administered surveys: Is it really a form of satisficing? Public Opinion Quarterly, 78(3), 591–633.
3. McCambridge, J., Witton, J., & Elbourne, D. R. (2014). Systematic review of the Hawthorne effect: New concepts are needed to study research participation effects. Journal of Clinical Epidemiology, 67(3), 267–277.
4. Arendt, S., Strohbehn, C., & Jun, J. (2015). Motivators and barriers to safe food practices: Observation and interview. Food Protection Trends, 35(5), 365–376.
