Estimate baseline sales and incremental sales

I need to set up a model for estimating baseline sales of beverages in order to determine the effect of different events on total sales. Available information is sales per product, day and store. The events that can be measured are price changes and distribution changes and in part out of stock situations. Unknown events are marketing activities, and it is these and pricing effects I want to identify and quantify. Information on product characteristics such as price, alcohol by volume, pack, size, several beverage classifications etc are available for grouping of products into comparable groups. The model will need to live on AWS, preferably calculated on current db instances (mysql, postgres or athena) as part of daily or weekly ETL. Data amount is <1m rows added per day (combination of date, product store).