Background: Hypophosphatemia is a condition that affects feed intake and milk production in dairy cows, and, to assess it, a precise biological parameter is needed. Plasma inorganic phosphate (Pi) can be used as a potential indicator of bioavailability of feed phosphorus in dairy cows. Objective: The purpose of this study was to evaluate the effects of sample type (serum vs. plasma), blood sampling site (jugular vs. coccygeal), and morning feeding on blood Pi concentration in dairy cows. Methods: Blood samples from 28 clinically healthy multiparous Holstein cows were analyzed for Pi concentration. The animals were sampled twice by coccygeal vessel into Vacutainer tubes both with and without sodium heparin, 3 hours after the morning feeding. In 4 additional cows, the concentration of Pi was measured in blood samples obtained simultaneously by jugular and coccygeal venipuncture into heparinized tubes before the morning feeding and 2, 4, and 6 hours after the feeding. The results were analyzed using parametric statistical tests. Results: Serum and plasma concentrations of Pi were correlated (r2= 0.999, P < .0001), but serum Pi concentration was higher than plasma Pi (P < .02). Plasma Pi concentration 2 hours after feeding tended to be higher than that before feeding in jugular samples, but the difference was not significant. There was no significant variation in plasma Pi during the morning postfeeding times, although jugular concentrations were 19% lower than coccygeal (P= .04). Conclusion: To increase the precision of plasma Pi measurements as an indicator of bioavailability of feed phosphorus, sample type, sampling time, and site must be taken into account, to correctly interpret plasma Pi concentrations.