I'd suggest matching after the unique id...

This works, but is oogly.

#!/usr/bin/perl
@chumptastic = <DATA>;
$chumpstain = join "\n", @chumptastic;
$chumpstain =~ m/bakjob2_details
                 .+?
                 credit
                 .+? 
                     # The ? makes these minimal matches; 
                     # thus catching the nearest match.
                 customer2\ =\ (-?\d+\.?\d*),/sx;
                 # Modified s to treat it as one string 
                 # (. matches newlines)
print "$1\n";
__DATA__
# Your Data goes here
[download]

Compressed, the regex looks like this: /bakjob2_details.+?credit.+?customer2 = (-?\d+\.?\d*),/s