Introduction: Dominant frequency (DF) analysis of atrial electrograms has been used to characterize atrial fibrillation (AF). The aim of this study was to explore technical issues that may affect the estimation of local activation rate during AF using DF analysis. Methods and Results: Epicardial atrial electrograms recorded during AF from 10 dogs were used to evaluate the effects of unipolar versus bipolar recordings, bipolar electrode spacing, postrecording processing, far field ventricular depolarizations, ventricular template subtraction, and signal duration on DF analysis. Simulated electrograms were used to evaluate the effect of far field ventricular depolarizations and signal‐to‐noise ratio. DFs were compared with activation rates obtained by manual marking and the reproducibility of the DFs was evaluated. Bipolar electrograms were found to be preferable to unipolar electrograms. Preprocessing was a necessary step for bipolar signals, but also aided analysis of unipolar recordings. Ventricular far field depolarizations significantly affected DFs. Ventricular template subtraction helped DF analysis in signals with both minimal and significant ventricular components. A recording duration above 2 seconds was required for reliable DF measurements. Signal‐to‐noise ratios below 13 dB could also affect DF, particularly for signals with significant amplitude and frequency variation. Conclusions: Various factors affect DF analysis. Proper interpretation of DF analysis requires careful evaluation of the AF signals and robust processing techniques.