I have 5 columns:
Quarter, Item, Type, Count, Date
I am trying to pull a UNIQUE count of date for each Quarter & Item combination ie. FY20Q3-AU has 2 different dates, FW20Q3-GLW has 1 unique date, FY20Q3-GLE also has only 1 unique date
Does anyone have any idea how I can accomplish this? I have been trying for a few hours modifying formulas that I found online without success.
If possible, I am trying to do this without an array formula (not sure if that is even possible)
I have tried:
{=SUM(--(FREQUENCY(IF(A:A=A2,MATCH(B:B,B:B,0)),ROW(B:B)-ROW(B2)+1)>0))}
| Quarter | Pricelist | Change Type | Item Count | Date |
|---------|-----------|-------------|------------|---------|
| FY20Q3 | AU | Type 1 | 1 | 4/18/20 |
| FY20Q3 | AU | Type 1 | 1220 | 4/4/20 |
| FY20Q3 | CH | Type 2 | 2 | 4/11/20 |
| FY20Q3 | CH | Type 1 | 1378 | 4/4/20 |
| FY20Q3 | GLA | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLA | Type 1 | 1421 | 4/4/20 |
| FY20Q3 | GLB | Type 1 | 1250 | 4/4/20 |
| FY20Q3 | GLB | Type 2 | 1 | 4/4/20 |
| FY20Q3 | GLC | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLC | Type 1 | 1404 | 4/4/20 |
| FY20Q3 | GLH | Type 2 | 2 | 4/11/20 |
| FY20Q3 | GLH | Type 1 | 1387 | 4/4/20 |
| FY20Q3 | GLME | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLME | Type 1 | 1421 | 4/4/20 |
| FY20Q3 | GLEE | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLEE | Type 1 | 1227 | 4/4/20 |
| FY20Q3 | GLEU | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLO | Type 1 | 211 | 4/4/20 |
| FY20Q3 | GLK | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLC | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLW | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLU | Type 1 | 1 | 4/18/20 |
| FY20Q3 | GLZ | Type 1 | 1 | 4/18/20 |
| FY20Q3 | JP | Type 1 | 1 | 4/18/20 |
| FY20Q3 | NHI | Type 1 | 1 | 4/18/20 |
| FY20Q3 | NHI | Type 1 | 1 | 4/18/20 |
| FY20Q3 | WCH | Type 2 | 2 | 4/11/20 |
| FY20Q3 | WCH | Type 2 | 2 | 4/11/20 |
So since you got access to O365 and it's DA-functions you would not need a CSE entered formula nomore. You can utilize UNIQUE and FILTER. For example:
=COUNTA(UNIQUE(FILTER(E:E,(A:A=A2)*(B:B=B2))))
Where I would suggest you won't use whole column references to not break down performance too much.