Three studies are reported in which the giving of information to customers, orally, in community pharmacies was tape recorded, and the customers were interviewed the following day. Customers' recall of information, including instructions considered to be essential, was low. In study 1, involving 44 customers in two pharmacies, a mean of 11 items of information was given to customers, but on average only three were recalled. Customers who took a more active part in discussion with the pharmacist (who were more likely to purchase self‐medication than to be prescription customers) tended to be given more new information and more repeated information, to recall more but also to forget more than passive customers. In study 2, 42 customers of three pharmacists were interviewed to determine the influence of a number of variables on recall. Whether customers reported feeling ill or in a hurry, or reported having received the information already from their general practitioner, or having had prior experience of the medication, were all unrelated to the amount of information recalled. The few customers who reported that the information provided by the pharmacist was not welcome were particularly poor at recalling it. In Study 3, the amount of information given or recalled was found to be unaffected by whether customers talked to a pharmacist in the busy sales area (n=25) or in a quiet counselling area (n=27). If pharmacists are to be effective providers of information about health and illness, attention must be paid to ways of improving customers' retention of information give