A comparison with a missing value is itself missing: NA > 5 is NA, and so is NA == NA. filter() keeps only rows where the condition is TRUE, so it drops rows where the condition is NA without a warning.
That causes two common mistakes. filter(discount == NA) returns nothing, because the test is never TRUE. And filter(discount != 0.1) looks as if it keeps "everything else", but it also loses every row with no discount recorded.
Use is.na(discount) to find missing values and !is.na(discount) to keep known ones. When you want missing rows kept alongside others, say so: filter(discount < 0.1 | is.na(discount)). Count rows before and after a filter when the numbers matter.
Find missing values, drop them, or keep them alongside other rows on purpose.