Caffeine and caffeinated coffee (CC) elicit acute insulin insensitivity when ingested before a carbohydrate load. The effects of CC on glucose tolerance and insulin sensitivity when co-ingested with a high carbohydrate meal and on postprandial metabolism of a subsequent (second) carbohydrate load have not been studied. In a randomised, crossover design, ten healthy males ingested either CC (5 mg caffeine/kg body weight), decaffeinated coffee (DC) or water (W; equal volume) co-ingested with a high glycaemic index cereal followed 3 h later by a 75 g oral glucose tolerance test. After the initial meal, insulin area under the curve (AUC) and insulin sensitivity index did not differ between treatments, although glucose AUC for CC (107 (sem 18) mmol/l × 3 h) and DC (74 (sem 15) mmol/l × 3 h) was greater than W ( − 0·2 (sem 29) mmol/l × 3 h, P < 0·05). After the second carbohydrate load, insulin AUC for CC was 49 % and 57 % greater (P < 0·01) than for DC and W, respectively. Despite the greater insulin response, glucose AUC for CC (217 (sem 24) mmol/l × 2 h) was greater than both DC (126 (sem 11) mmol/l × 2 h, P = 0·01) and W (55 (sem 34) mmol/l × 2 h, P < 0·001). Insulin sensitivity index after the second meal was lower after CC (8·2 (sem 0·9)) compared with both DC (12·4 (sem 1·2), P < 0·01) and W (13·4 (sem 1·4), P < 0·001). Co-ingestion of CC with one meal resulted in insulin insensitivity during the postprandial phase of a second meal in the absence of further CC ingestion. Thus, CC may play a role in daily glycaemic management.