Second Try

Imagine the following table:

Datahour,A,B,C,D,E,F

You get a new record everytime something happens.

In DataHour 11, you get substantially more things happening than happened in DataHour 10.

So, you want to figure out which particular data point might be bulking up to give you more records.

So, you effectively do:

SELECT datahour, a, COUNT(*)

FROM table

GROUP BY datahour, a

for each field A, B, C, D, E and F, hoping to find the particular field where it went from low to high.

There's got to be a term for this type of data analysis, but I can't figure out what it is.

Forget the old website example.  This isn't particular to websites.  Or emu births.  Or lottery tickets sold.  Fields A-F could just as easily be 'Client IP, UserAgent, CountryCode, etc' as it could be 'Weight, Height, Eye Color, Hair Color, Payment Type, etc'.