Compared to the 1990s when HbA1c was established as a marker for glycemic control, the accuracy of the HbA1c assays has greatly improved and HbA1c is currently used also for the diagnosis of type 2 diabetes. For most patients, the agreement is excellent between glycemic status and HbA1c results achieved by various methods. However, for patients with increased erythrocyte turnover, HbA1c does not reflect the glycemic status. For patients with rare haemoglobin variants the HbA1c value might be falsely decreased or increased, depending on which HbA1c assay has been used. Recently, falsely increased HbA1c results due to aspirin interference in an ion-exchange method were reported. When misleading results are suspected comparison of results from different HbA1c methods using different analytical principles or continuous glucose monitoring might be useful. HbA1c is likely to remain the most important marker for glycemic control, but complementary tests have to be established.