Methods, apparatus, systems and articles of manufacture are disclosed to calculate long-term effects of marketing campaigns. An example method includes for respective participants, identifying, with a processor, a trial purchase of a first product associated with a brand of interest following a marketing stimulus for the brand of interest, for the respective participants, incrementing, with the processor, a brand purchase count of purchase occasions when a subsequent product associated with the brand of interest is purchased during an exposure period of the marketing stimulus, for the respective participants, resetting, with the processor, the brand purchase count to zero when a subsequent product unassociated with the brand of interest is purchased during the exposure period of the marketing stimulus, generating, with the processor, purchase groups indicative of consecutive purchase occasion values of the respective participants based on respective highest values of the brand purchase count, and calculating, with the processor, the sales effect for the respective purchase groups based on the respective values of the brand purchase count.