Count Rows where not in Previous Period

[leahschneider]

I have a visual that sums total sales for a product category over time. For each period (usually month), I would like to sum the total number of product categories that aren't in the previous period. 

Example Raw Data:

Date	Category	Product	Total
31-Jan	Electronics	Watch	43
31-Jan	Clothes	Shirt	254
31-Jan	Clothes	Hat	478
31-Jan	Misc	Giftcard	2
28-Feb	Gear	Bike	24
28-Feb	Gear	Helmet	3
28-Feb	Electronics	Watch	25
28-Feb	Clothes	Shirt	576
31-Mar	Gear	Bike	15
31-Mar	Electronics	Watch	3
31-Mar	Misc	Giftcard	1

Desired Results:

Date	New Category Count	Explanation
31-Jan		The first date doesn't matter. It can also say 4.
28-Feb	1	Gear is not in January products.
31-Mar	1	Misc is not in February products. It is in January, but that should not affect this calculation.

I do not know why this doesn't let me make tables that are usable looking. This is a picture of what I have and what I want:

[CNENFRNL]

It's, in essence, nothing but an question of MoM comparison.

[leahschneider]

Thank you this is very helpful. I had wanted count and not concat but that's pretty easy to fix. My only issue with this is that it's not possible to group up. If I want to add quarter or year to the table, it won't work correctly on quarter or year.