We tested the hypothesis that swabs from the nasopharynx carry a higher viral load than swabs from the oropharynx in patients with real-time polymerase chain reaction (PCR)-confirmed influenza infection. Using flocked swabs, oropharyngeal and nasopharyngeal samples were harvested from hospital-admitted influenza patients no later than 3 days after the initial detection of influenza virus. Comparison of cycle threshold (CT) values was performed to assess differences in viral load in the specimens. Seventeen patients were diagnosed with influenza B, 14 patients with influenza A(H1N1)pdm09, and one patient with influenza A(H3N2). Nasopharyngeal samples were positive at a lower CT value than the oropharyngeal samples [mean difference in CT 5.75, 95 % confidence interval (CI) 3.8-7.7, p < 0.01], suggesting that, on average, the calculated viral load of the nasopharyngeal samples was 54 times higher (95 % CI 13.7-210.8) than those of the oropharyngeal samples. The corresponding difference in the calculated viral load for influenza A(H1N1)pdm09 virus was 23 times (95 % CI 3.8-136.2, p < 0.01) and for influenza B virus, it was 80 times (95 % CI 9.3-694.6, p < 0.01). In patients with acute influenza, nasopharyngeal swabbing was clearly superior to oropharyngeal swabbing in terms of diagnostic yield by real-time PCR.