How to count amended data without repeats

Can anyone help me, I am trying to understand some key info on a very large spreadsheet (image below).

It is all about an order review system- there are three main columns for me: A. order number. B. Initial order type and C.Final order type.

I want to know how many orders that come in as Type 1 are changed to type 2, and 3. And vice-versa: how many are changed to Type 1 from 2 or 3. (Type 1 is my main concern here).

The problem is, before now, every-time any change was made to an order, this order was reproduced onto a new row in the sheet (when changes may not have happened to the actual order type, but rather the details). Another issue is that a lot of orders come in with the type as blank in column B. This blank order is then assigned a type in column C (so does not count as having changed).

I need to know:
-How many individual orders there are in total (some orders won't change of course)
-How many individual Type 1 jobs were recorded at the beginning of the process (column B)
-How many individual Type 1 jobs were recorded at the end of the process (column C)
-How many individual jobs changed from type 1, to type 2 or 3
-As well as how many changed from Type 2/3 to type 1.
Not counting repeats or blanks.

I started doing this with filters (see pic) and a count of how many have been changed (excluding blanks), then how many began as type 1 and became a different type, and how many began as type 2/3 and ended up as type 1, but I keep being tripped up by repeats. I also tried simply removing duplicates, but then it indiscriminately removes repeats and can remove the repeated list item that shows the type change. I have tried for a long time on this and have reached my limit of understanding; are there any excel braniacs out there who can share a simple solution?

Screenshot 2020-04-30 at 02.45.00.png

Are you REALLY using the latest Excel version, which would be from Office 365? If so, you have new worksheet functions FILTER and UNIQUE. Use them.

Distinct order numbers
=ROWS(UNIQUE($A$2:$A$100001))

Distinct order numbers initially Type 1
=ROWS(UNIQUE(FILTER($A$2:$A$100001,$B$2:$B$100001="Type 1")))

Distinct order numbers ultimately Type 1
=ROWS(UNIQUE(FILTER($A$2:$A$100001,$C$2:$C$100001="Type 1")))

Distinct order numbers initially Type 1 which become Type 2/3
=ROWS(UNIQUE(FILTER($A$2:$A$100001,($B$2:$B$100001="Type 1")*(($C$2:$C$100001="Type 2")+($C$2:$C$100001="Type 3")))))

Distinct order numbers initially Type 2/3 which become Type 1
=ROWS(UNIQUE(FILTER($A$2:$A$100001,($C$2:$C$100001="Type 1")*(($B$2:$B$100001="Type 2")+($B$2:$B$100001="Type 3")))))

If you're using Office 2019 (there is no Office 2020 AFAIK), there's no FILTER or UNIQUE. How is your data sorted when there are multiple instances of the same order number in column A?

Thank you so much for this; I do have Office 365, and those seem to work; I have two questions though, how can I make sure an array works on my mac (recent change, know on a PC you press shift enter)

and, the first one gives a different number than pressing 'remove duplicates' from column A (in the data tab); why is this?

Thanks so much for your help; this is a huge weight off my mind!

Hi! Please see attached sample, with the arrays suggested below