StartPoints in custom fitting models

Hi, The method of NonlinearLeastSquares for fitting experimental data has an option for StartPoint for each coefficient to be determined. However, with the same data and equation, the StartPoint changes and there is a variation in the values of coefficients obtained. Hence, I would like to know the reason for the changes in StartPoints even with the same data and model along with procedure to eliminate it to obtain unchanging coefficient values from the fit model. Thank you.