CMV infection is a major challenge in allogeneic stem cell transplantation (allo-SCT). The changing landscape in CMV management includes the introduction of letermovir in prophylaxis of high-risk patients and the source of CMV DNA monitoring (plasma-PL vs. whole blood-WB), for pre-emptive therapy (PET) initiation. We report here how our real-life experience in CMV management evolved, following letermovir registration. We focus on: (i) the effects of systematic use of letermovir for CMV prophylaxis in high-risk patients, (ii) the results of a longitudinal comparison of CMV DNAemia monitoring in PL and WB. From December 2018 to April 2020, 60 allo-SCTs have been performed in our center (LET ERA), of whom 45 received letermovir in prophylaxis from day 0 to day + 100, because of recipient positivity of anti CMV IgG. These patients were compared with a cohort of 41 allo-SCTs performed between November 2017 and November 2018 (NO LET ERA). Firstly, the incidence of CMV clinically significant infections, CMV disease, bacterial infections, proven/probable fungal infections, hospital re-admissions after allo-SCT by day + 100 in the two ERA were 8 vs. 44% (p = 0.0006), 2 vs. 12% (p = 0.02), 37 vs. 56% (p = 0.05), 8 vs. 19% (p = 0.09), and 23 vs. 39% (p = 0.09), respectively. By day + 180 these differences were 17 vs. 68% (p < 0.00001), 2 vs. 12% (p = 0.02), 45 vs. 78% (p = 0.09), 8 vs. 22% (p = 0.05), and 40 vs. 66% (p = 0.01), respectively. Secondly, from February to May 2019, we comparatively measured CMV DNA from WB and PL and we confirmed that there is a linear correlation between CMV DNA level in WB and PL (Spearman's test r = 0.86). Moreover, CMV DNAemia at the time of PET in the 12 patients with a clinically significant CMV infection was higher in WB vs. PL (5.202 vs. 4.981 copies/ml, p = 0.1). Our real-life experience confirms that: (i) letermovir is highly effective, leading to a significant drop in CMV clinically significant infections and CMV-related complications by day + 100 and + 180 after allo-SCT; (ii) WB may be an effective alternative to PL as a source for CMV DNA monitoring, as a linear correlation of DNAemia was confirmed between WB and PL, even if the CMV DNAemia at PET initiation was comparable in the two sources.