Offices & Operations mentor Charles Bodenstab responds: The answer to this question is directly related to the nature of your business.
In my experience, monthly data is sufficient for most business situations -- particularly when you realize that the statistical forecasting techniques blend or average the data together anyway. There is a tendency to think that the finer the data is sliced, the more useful it will be. But this is not so. In most cases, I'd suggest sticking with monthly data unless issues such as these apply.
If the nature of your business is such that there are significant patterns of demand within the month that need to be taken into account, then weekly data starts to make sense. For example, if you are in retail, and Easter is a key holiday, you will have problems if you don't track weekly data and incorporate a seasonality adjustment.
Additionally, if your sales data has a distinct and consistent pattern within the month and your vendor lead times and reorder cycles are in the one- to two-week range, then weekly data and weekly forecasts will enhance your inventory performance.