Pavement performance monitoring is an essential part of a pavement management system (PMS). Therefore, highway agencies collect pavement condition data containing various structural and functional distresses on a regular basis. How frequently pavement condition data are collected varies. The effect of pavement monitoring frequency on pavement performance prediction and its consequence at network-level PMS decisions were explored. A statistical methodology was developed to investigate the impact of monitoring frequency on performance prediction with different model forms. The results of the analyses showed that performance predictions were affected by monitoring interval. Furthermore, the various types of distress to be collected might need different monitoring intervals because of their unique growth over time and associated uncertainty in prediction. When cost consequences and prediction uncertainties were combined, monitoring of cracking (image based) at 1-year intervals appeared to be more appropriate, whereas for roughness (sensor based), a monitoring interval of 1 to 2 years appeared to be suitable. The results of network-level analyses demonstrated that monitoring interval might significantly affect the short- and long-term network conditions for various preservation strategies. Increasing the monitoring interval may have consequences for PMS decisions. Longer intervals for crack monitoring (image based) might cause an overestimation of agency costs for pavement repair at the network level. Longer intervals for international roughness index monitoring (sensor based) will result in an underestimation of repair costs at the network level.