Freeze-drying is increasingly used to preserve both the viability and the acidifi- cation activity of lactic acid bacteria (LAB). However, their resistance to freeze-drying and to storage depends on the microorganisms themselves and is strongly influenced by the process and storage conditions. Freeze-drying is basically defined in three stages: freezing (solidification); primary drying (ice sublimation); and secondary drying (moisture desorption). The freezing step is of paramount importance in freeze-drying. It governs ice crystal morphology and size distribution, which in turn influence several critical parameters, including the stability of some proteins, the primary drying rate, and the secondary drying rate. Although, there is some reported work on the effect of freezing kinetics on LAB cryopreservation, there is a lack of information on the impact of the freezing step on LAB activity recovery after freeze-drying. The aim of this study was to investigate the impact of various freezing methods on the stability of lactic acid bacteria after freeze-drying and storage at ambient temperature. Studies were carried out on Lactobacillus delbrueckii subsp. bulgaricus CFL1 protected with sucrose. Samples were frozen in vials using different freezing protocols including cooling at 1 C/min or 0.3 C/min, immersion in liquid nitrogen (140 C/ min), putting vials on the pre-cooled shelf (50 C), annealing treatment, controlled nucleation by Pseudomonas syringae followed by immersion in liquid nitrogen or by a slow cooling rate (1 C/min). After the freezing step samples were freeze-dried according to a standardized protocol and stored at ambient temperature. The LAB acidification activity and viability were determined before and after freezing, after freeze-drying and during storage by using the CINAC system and plate counts, respectively. In order to relate the activity and viability of freeze-dried LAB to the ice crystal size and distribution obtained during freezing, vials were examined following freezedrying by using scanning electron microscope (SEM). (Conflicts of interest: JM is a shareholder in Asymptote Ltd. Source of funding: None declared.)