Selected electrocardiograph (ECG) intervals may be useful when incorporated into prediction models for cardiac risk assessment in dogs. Standard recommendations for ECG acquisition may not be adhered to in practice. Study objectives were to compare duration of P, PR, QRS, QT, and R wave peak time intervals in: (1) lead II ECGs vs. single lead precordial ECGs in conscious dogs; (2) lead II ECGs with electrodes placed in proximal limb (PL) vs. distal limb (DL) positions with dogs in right lateral (RL) recumbency, left lateral (LL) recumbency and standing positions; (3) single lead precordial ECGs from dogs in RL, LL and standing positions; and (4) before and after sedation with butorphanol in lead II ECGs obtained in RL recumbency. All intervals could be measured in all dogs (conscious and sedated) from a RL lead II ECG with both PL and DL electrode positioning. This was reduced to 98% for lead II ECGs with dogs in LL and standing positions. Intervals that were not different regardless of recording device, dog position, electrode limb position or sedation included P, QRS and P + QRS, suggesting that these intervals have the greatest clinical utility across a variety of recording conditions. The main impact of positioning in healthy dogs was the lack of ability to consistently measure all intervals in standing dogs, particularly P wave duration. Further investigation is needed to determine if this is applicable to dogs with cardiac disease.