PowerPivot measures work with aggregates. For instance, you can’t define a measure with a formula of =SalesTable[Amount Column] – you need to have an aggregation function or you will get an error. So SUM(SalesTable[Amount Column]) is legal.

Strictly speaking I did not need to define an [OpenDate] measure. I could have just used LASTDATE(Stores[OpenDate]) everywhere in my other measures where I referenced [OpenDate]. But I needed an aggregation function (LASTDATE in this example) to make my formulas legal. At that point I might as well just define it once as a measure and re-use it.

I’m new to PowerPivot so forgive me for this basic question. I was just wondering why we need to create the measure for Open Date as the first step if it’s just going to pull the date directly from the existing Open Date column? Is there something special about creating a measure that allows you to use it in these various formulas where a direct reference to a column does not allow this?

[Qualifying Store Count] =

CALCULATE(COUNTROWS(TheNewTable),
FILTER(TheNewTable,
TheNewTable[StartDate] < = FIRSTDATE(Calendar[Date]) -365 &&
TheNewTable[EndDate] >= LASTDATE(Calendar[Date]
)
)

and then in your FILTER for the SSS measures themselves, do something like:

FILTER(Stores, [Qualifying Store Count] > 0)

1) Create a new table in the model called StoreActiveDates (or similar)
2) This table has three columns – StoreID, StartDate, EndDate
3) Unlike the Stores table though, this table can contain MULTIPLE rows for a single store
4) Relate this new table to the Stores table, using the StoreID column

Now write a measure (or measures) on the StoreActiveDates table that answers the question “Should this store count?” It might be a CALCULATE(COUNTROWS(), expression…) type of thing.

Lastly, change the filter expression in the FILTER function of the measures covered here to use THAT measure rather than the [Is Store Closed] measure for instance.

What I meant to say was: yes, I think that is a very good fit for many business situations. I love this about PowerPivot – once we get started down a road, all kinds of options open up. And since the Excel Pros working on this stuff tend to be tightly embedded into the “business side of the house,” we are properly positioned to come up with precisely the right ideas.

Glad to see a post dealing with SSS on retail business
I’m working in that business as financial controler for a while, and since day one, I struggled to set up somthing automatic to calculation SSS.
I managed through different excel formulas, until powerpivot has been created and released.
During years, I was found of the “did we have sales” approach. Makes more sense to me. But as said, every business situation is different. Now some folks want to have a similar approach as described by Cory.
The only difference with situation described by Cory are :
- are included on SSS only shops that are opened 12 months without discontinuing on both 2 fiscal years (a fiscal year here starts in July 1st Y and finishes June 30th Y+1)
- some stores are opened during two successive fiscal year, but are somewhen during a year, closed for refit. And so naturally they should be excluded from like for like comparison. So compared to the fictional open/close store table you’ve created here, one shop can have several opening and closing dates.

To manage that, so far I created a list of stores that are to be included in SSS, and maintain that list month after month (depending on closing / opening). But I’m keen to automatise it a little bit more.

If you have any idea, I’ll be very grateful, and pretty sure a lot of folks reading your blog will be happy to learn that.

Have a nice day

Regards