Is it better to introduce seasonal factors to multiply rather than add up

Yes. The additive model assumes that the effects of seasonal factors on the development of a phenomenon are independent of those of the long-term trend component, the cyclical variation component and the irregular variation component and are measured in the same units as the values of the variables in the time series. The multiplicative model, on the other hand, assumes that the effects of seasonal and other factors are reciprocal and are expressed as ratios (relative quantities). In a time series, not all four types of factors are present at the same time in every time series. However, in general, long-term trends are always present, while seasonal and cyclical factors are not necessarily present. When the seasonal and cyclical components are absent, S or C in the multiplicative model takes the value of 1, and in the additive model S or C takes the value of 0.

The additive model assumes that the seasonal and long-term trend components, the cyclical and irregular components are independent of each other in terms of their effects on the development of the phenomenon, and that the units of measurement are the same as the units of the values of the variables in the time series. The multiplicative model assumes that the effects of seasonal and other factors are reciprocal and are expressed as ratios (relative quantities). In a time series, not all four types of factors are present at the same time in every time series. In general, however, long-term trends are often present, while seasonal and cyclical factors are not. When seasonal or cyclical components do not exist, S or C takes the value of 1 in the multiplicative model and 0 in the additive model.