Conditional Summing of the Top 80% of Data

All...I NEED AN EXCEL GENIUS, PLEASE!!! Looking for some help in Excel 2010 and what appears to be a very specific scenario. I have a data set that reflects sales for a specific category of items. Of the entire data set, I need to get to a total of the sales for only the top 80% of that group. Gets trickier because I then need to count that group to understand how many data elements make up the 80%. I see this as two separate formulas...data example is as follows

(Column A) (Column B)
Item ID Total Sales
Item 1 $100,000
Item 2 $5,000
Item 3 $40,000
Item 4 $10,000
Item 5 $6,000
Item 6 $10,000
Item 7 $8,000
Item 8 $4,000
Item 9 $500
Item 10 $3,000

So for the above, I need one formula that says what the total revenue is for the top 80% of the items without exceeding the 80% threshold. So for the above, the formula should render a total of $140,000 and then I would need another formula that says the count of items to get this is 2. Make sense? I can't for the life of me get something to work. Please help!!! Thanks in advance!!!

Thanks for your prompt inquiry. Unfortunately, in the actual data set, sorting is not ideal b/c of the size and various sub groups that I have within it.....is there a way to solve this issue with out sorting?

Solution from martin wilson, without helper column:

Please Login or Register  to view this content.

Confirmed with Ctrl-Shift-Enter

Thanks for everyone's responses. Really appreciate everyone trying to lend a hand. Sadly, I'm still struggling to get any of these solutions to work (I should note that I am not an Excel genius like yourselves) I should have clarified that my data is not sorted so I will have multiple subcategories which need to be isolated and then have the top 80% calculations applied (totaling of the top 80% of sales for each subcategory and then the count of items that make up that 80%). Richard, your code is most intriguing and would be great to have a built in formula that gets to both...but unfortunately can't get it to work. It returns a #value! error when I attempt to use the formula. Any thoughts?

Martin/Bebo, the percentrank formula is challenging because of the multiple subcategories within the larger population set. I can't seem to get this work for me either as it would have to be a conditional percentrank after identifying the subcategory. Assuming I could get this to work, I'm still challenged to do the count to identify those items that make up the top 80% in each categories.

Please keep the suggestions coming...IT IS VERY MUCH APPRECIATED. thanks!!!

RICHARD YOU ARE A LIFE SAVER and A GENIUS. This is EXACTLY the kind of formula I'm looking for....I hate to ask for one more favor and this is totally on me for not clarifying up front, but is there anyway to make this "conditional" friendly...similar to the SUMIF function, is there a way to leverage the function to select only those items with a certain criteria and perform this calculation. I'm a proposing a TOPIF kinda of function....for example, using the same data set from before if I add in another descriptive column...such as...

(Column A) (Column B) (Column C)
Item ID Description Total Sales
Item 1 Hat $100,000
Item 2 Shirt $5,000
Item 3 Shirt $40,000
Item 4 Tie $10,000
Item 5 Socks $6,000
Item 6 Shirt $10,000
Item 7 Shirt $8,000
Item 8 Socks $4,000
Item 9 Hat $500
Item 10 Socks $3,000

....and I wanted to perform the "Top" function on only "shirt" group for example, is there a way to allow such criteria to be defined and have the formula adjust accordingly. I have about 15000 rows of data in total and need to segment this down to get to the top 80% sum and counts for each grouping. I apologize for imposing on you further but you are clearly very good at what you do and if this is possible, it would be HUGE for my analysis.

Thank you so much!!!