A         B     C
 1  NA        NA    NA
 2  NA       200    NA


              A        B       C
1  NotImportant       NA      NA
2       Unknown      200      NA

if(is.na(ColumnA)) ColumnB[is.na(ColumnB)] <= "Not Important"
if(is.na(ColumnA)) ColumnB[!is.na(ColumnB)] <= "Unknown"

df$A <- c("Unknown", "NotImportant")[(rowSums(is.na(df)) == ncol(df)) + 1L]
df
#              A   B  C
# 1 NotImportant  NA NA
# 2      Unknown 200 NA

df <- structure(list(A = c(NA, NA), B = c(NA, 200L), C = c(NA, NA)), .Names = c("A",
"B", "C"), class = "data.frame", row.names = c("1", "2"))

df[rowSums(is.na(df))==3,]$A <- "NotImportant"
df[with(df, is.na(A) & (!is.na(B) | !is.na(C))),]$A <- "Unknown"

#             A   B  C
#1 NotImportant  NA NA
#2      Unknown 200 NA

df <- structure(list(A = c(NA, NA), B = c(NA, 200L), C = c(NA, NA)), .Names = c("A",
"B", "C"), class = "data.frame", row.names = c("1", "2"))

library(dplyr)
df%>%
mutate(A=case_when(
 is.na(.$A)&is.na(.$B)&is.na(.$C)   ~ "Not Important",
 is.na(.$A)&(is.na(.$B)|is.na(.$C)) ~ "Unknown")
 )

              A   B  C
1 Not Important  NA NA
2       Unknown 200 NA