The objective of this observational study was to estimate partial direct costs of dry cow antibiotic therapy (DCT) protocols used on 37 large dairy herds in Wisconsin and to estimate the potential monetary savings and reduction in antimicrobial usage (AMU) if selective DCT was used. Partial direct costs of DCT were calculated using costs of intramammary (IMM) antimicrobials and teat sealants. Data were retrospectively collected on 37 large dairy farms for a period of one year and included the total number of cows dried-off, types of IMM antimicrobial used, and product prices (obtained from farm invoices). A single farm visit was performed to verify data. Clinical mastitis (CM) and SCC history across lactation were used as the criteria to identify cows eligible to receive only teat sealants (no antibiotic DCT) based on adoption of a hypothetical selective DCT program. Descriptive statistics were performed using PROC MEANS to summarize continuous herd and cow characteristics. Differences in costs among IMM antibiotic products and eligibility to not receive antibiotics at dry off by parity were analyzed using ANOVA. Differences in milk yield at the last test day and DIM at dry-off based on eligibility for selective DCT were analyzed using PROC MIXED. A total of 35,691 cows were dried-off across all herds (n = 37) and most of the herds used IMM antibiotic DCT at dry-off in most of the cows. Teat sealant was used as part of the dry-off protocol in all but 3 herds. Of the enrolled farms, 30 used blanket antibiotic DCT in all quarters of all cows and 7 herds used selective DCT with no administration of antibiotics given to 0.8% to 58% of the dry cows within herd. Across all farms, the average cost of products given per dried cow was $19.57 ± 0.64 and ranged from $8.72 to $24.04. As expected, the cost per dried cow was greater ($23.45 ± 0.38) in herds that used higher-cost intramammary antibiotics as compared with herds that used lower-cost intramammary antibiotics ($16.64 ± 0.40). When using an algorithm based on udder health records to hypothetically select cows that would be eligible to receive teat sealants only (no antibiotic DCT), eligibility ranged from 27.3% to 93.3% within-herds and varied by parity, milk yield at last test day, and DIM at dry-off. If a selective DCT program based on udder health records was used, an overall reduction of $5.37 (27% of total costs) per dry cow would be expected. Likewise, adoption of selective DCT based on udder health records would reduce AMU at dry off by approximately 51%. However, neither of these estimates included potential adverse health complications (such as increased clinical mastitis) that can occur if selective DCT programs are not effectively performed. Variations in partial direct costs at dry-off were observed among herds based on treatment protocols. In herds that have good udder health management and contagious mastitis pathogens controlled, adoption of selective DCT based on SCC and history of clinical mastitis can reduce AMU and partial direct costs of dry cow management.