Snow-farming is one of the adaptive strategies used to face the snow deficit in ski resorts. We studied the impact of a shifting snow-farming technique on a pasture slope in Adelboden, Switzerland. Specifically, we compared plots covered by a compressed snow pile for 1.5, 2.5 or 3.5 years, which then recovered from the snow cover for three, two or one vegetation seasons, respectively, with control plots situated around the snow pile. In plots with > 1.5 years of compressed snow pile, plant mortality was high, recovery of vegetation was very slow, and few plant species recolonized the bare surface. Soil biological activity decreased persistently under prolonged snow cover, as indicated by reduced soil respiration. The prolonged absence of fresh plant litter and root exudates led to carbon (C) limitation for soil microbial respiration, which resulted in a significant decrease in the ratio of total organic carbon to total nitrogen (TOC/TN) under the snow pile. Microbial C, nitrogen (N) and phosphorus (P) immobilization decreased, while dissolved N concentration increased with compressed snow cover. Longer snow cover and a subsequent shorter recovery period led to higher microbial C/P and N/P but lower microbial C/N. Nitrate and ammonium were released massively once the biological activity resumed after snow clearance and soil aeration. The soil microbial community composition persistently shifted towards oxygen-limited microbes with prolonged compressed snow cover. This shift reflected declines in the abundance of sensitive microorganisms, such as plant-associated symbionts, due to plant mortality or root die-off. In parallel, resistant taxa that benefit from environmental changes increased, including facultative anaerobic bacteria (Bacteroidota, Chloroflexota), obligate anaerobes (Euryarchaeota), and saprophytic plant degraders. We recommend keeping snow piles in the same spot year after year to minimize the area of the impacted soil surface and plan from the beginning soil and ecosystem restoration measures.