A Pareto analysis asks how few products make most of the sales. For each product, add up the sales of every product that sells at least as much, and divide by the total. That's its cumulative share:
Cumulative Share =
VAR Cur = [Total Sales]
VAR Products = ADDCOLUMNS(ALL(Product[Product]), "S", [Total Sales])
RETURN DIVIDE(SUMX(FILTER(Products, [S] >= Cur), [S]), SUMX(Products, [S]))
The best seller's cumulative share is its own share; the last product reaches 100%.
An ABC classification cuts that curve: A up to 80%, B up to 95%, C the rest. Buyers protect A items and question C items.
Edge cases: products tied at the cut-off land in the same class, and a product with no sales sits at 100%, deep in C.