Average over dates for benchmark

Hi there

I am trying to understand the difference in sales pre and post event to see if there is any change. 

I have payment files for 6 dates, in a single table

'SaleFiles'

With columns:

SalesValue,SaleName,MembershipDate,DateKey,Customer ID

Each customer ID may have multiple sales/refunds on the same day under different types of SaleName.

i.e:

£10,Membership,02/07/2022,44744,12345

£100,Ticket,02/07/2022,44744,12345

-£20,Refund,02/07/2022,44744,12345

I then have a date table:

The dates I have are: 7 May 2022, 4 June 2022, 02 July 2022, 30 July 2022, 27 Aug 2022, 24 Sep 2022

With Columns:

Date,DateKey

These are joined on the date key.

One (DateTable) to Many (SaleFiles)

What I want to be able to achieve is, what was the average value of Ticket sales over the 7th May & 4th June combined.
Then be able to put this into a table to show the difference between this average and each of the following dates.

I would hope to use, DateTable as my date in table or x-axis as this will filter other information elsewhere.

I would then also want to be able to drill through by customer to see if the value per customer has increased or decreased on each on after the 4th June.

I've tried the following but I can't seem to get it all to stick so I can see the difference.

It feels like this should be really simple but I've tweaked all the variations of FILTER | AND | ALL within the measure and for some reason it always ends up with a problem somewhere, either averaging every date seperately (not combining the first two) or returning blanks for any dates outside of the filter date.

**UPDATE**
I've got this far and the calculation works per customer until I try to display this over a date table.

Any help greatly appreciated.