Count amount of unique values with multiple conditions

Hi,

I can't succeed to count the amount of unique values with multiple criteria. The columns are employee#, week, year, date and invoicing method.
Employee#: the employee# occurs so now and then double within a date. Employee# should not be counted if there is more than one unique number on unique date.
Week: The cutoff date is variable and is set on the first of may for now (Week 18). The search has to be of last month (4 weeks) on date or in the week column, so < week 18 and > week 13
Year: Year of cutoff date, so 2015.
Date: Each row has a date. As stated above there are dates with duplicate employee#'s.
Invoicing method: Should only be counted when "YES"

There has been already a similar post about this issue but I can't seem to comprehend it and apply it on my sheet.

Anyone fancy helping out?
Please find attached an Example.xlsx.


Looking forward to the responses.
Edit: Right answer is 6.
The rows that apply to the criteria are:
5 (regular)
6 (or 7 since it is a duplicate with same date)
9 ( Same employ# but seperate date)
10 (Same employ# but seperate date)
12 (Different employ# same date)
13 (Different employ# same date)

I need a little clarification. Are you saying employee ID 5 shouldn't be counted since they're both on 4/6/15 but employee ID 7 should because they are on separate dates? I don't come up with 5 that's why I'm asking. I come up with 2 (if we're not counting #7) or 4 if we are counting #7.

Why are you including week 13? To me, 5/1 minus 4 weeks only goes back to 4/3. I'm only asking for clarification as I work through the math

Maybe with a helper column...

On my first effort, I grouped individual and weeknumber in the helper. It now transpires that it should have been individual and date (which would give the answer 6). The attached now does that...

Thanks. There are probably ways round it, but if it keeps a more manageable and easily maintained formula, is it worth it?? You can always just hide the column that it's in...

Fair enough comment about using cell references, and about susing a structured tabble. However, there's no reason not to use whole-column references with countifs; so the formula I suggested could just as efficiently be phrased as:=COUNTIFS(B:B,"=Yes",C:C,">13",C:C,"<18",F:F,"=1")

My suggestion is to definitely use table format. I have only begun to find the benefits of that format over the last few months and I'll never go back now!!!!

1) Pivots definitely work off tables. Actually, pivots work better on tables because you don't have to keep updating your data range with the "change data source" button as you add data. If it's in table format the range will update to the entire table whenever you click "refresh all"....I LOVE this feature as I hated updating ranges.

2) If you don't want to use the whole column as the range (I have run into some issues), once you format as a table, just update the range to the new table range. Now does that sentence make sense? No...here's an example using your data:
*change the data in A1:A13 to be formatted as a table
*go to the countifs formula in N6 and delete the B2:B18 for the first criteria range.
*highlight B1:B13...it will update in the fomula to say "Table1[[#All],[Invoicing method]]"
I did it in this attachment --- Example.xlsx


Benefits:
*"Table1[[#All],[Invoicing method]]" is like a named range...so no matter how many rows you add...as long as you add them into the table, the formula will keep updating
*the formula in column F will keep filling in as you add data
*use "refresh all" and your pivots will auto update after you add data (note: since your pivots are already set up, you might have to do "change data source" one time and set it to the table range. Then you should be good after that. Maybe that'll be unnecessary...I'm not sure.

I love tables!