Consistent deviations of the 3M Petrifilm aerobic counts (AC) from the standard pour plate aerobic plate count (APC) were observed with dehydrated onion and garlic products. A large study was designed to determine the relationship of these two methods and the root cause for the deviations. A total of 3,800 dehydrated onion and garlic samples were analyzed by both the Petrifilm AC and the standard pour plate APC method. Large spreader-like liquefied areas were observed on numerous Petrifilm plates. These liquefied areas made enumeration inaccurate. “Liquefier” microorganisms from Petrifilm plates were isolated and identified to species level by 16S rRNA and gyrB gene sequencing. Enzyme diffusion assay was performed to determine potential enzymatic degradation of guar gum, the gelling agent used in Petrifilm plates. The results indicated that the correlation between Petrifilm AC and standard APC is relatively low. Paired t test results suggested that the Petrifilm AC method produced significantly different results compared with standard APC. The discrepancies were attributable at least partly to a liquefier organism that hydrolyzed guar gum, leading to liquefaction. Liquefaction of Petrifilm plates seems to have two effects on accuracy: (i) liquefied areas may allow motile organisms to move and multiply in the liquefied area during the incubation period, yielding more than one colony from one cell and, as a result, leading to overestimation of the microbial load and (ii) the blurred areas obscure other colonies, leading to potential underestimation. The liquefier organism was identified as Bacillus amyloliquefaciens, a potent mannanase producer and heat-resistant spore former. Enzyme diffusion assay confirmed that mannanase contained in the cell-free supernatant of B. amyloliquefaciens can hydrolyze the 1,4-β-mannopyranosyl bond, the backbone of guar gum. This is the first report of the role of B. amyloliquefaciens in the liquefaction of Petrifilm plates and its negative impact on accuracy.