You could use an ORDER BY statement to order the data by Customer and Date. Next, create a field in the table to first check if the Customer is the same, then check if the value has changed. It would look something like IF(Customer <> PREVIOUS(Customer), 'N', IF(Value <> PREVIOUS(Value), 'Y', 'N')) as ValueHasChangedComparingPM. If the Customer has changed, then 'N' because that indicates the next Customer's cluster of Values (thanks to your ORDER BY statement), else if the value has changed then 'Y' because that indicates there's a changed in Value from one time period to the next (also thanks to your ORDER BY statement), else 'N' because that means the Values were the same from one time period to the next.
Something I forgot to mention... The ORDER BY statement only works on a RESIDENT load. So you would have to load it in from source first. Also, bear in mind this will remain static if you are applying other filters. To have it be dynamic based off table sorting or filters, I would use a similar formula as above, but in a table, variable, etc., and using ABOVE instead of PREVIOUS