Why do Web Analytics such as Google Analytics use Dimensions and Metrics instead of SQL statement?

Question

Just when I get quite familiar with SQL statements once again, when pulling data from Google Analytics, I found that they don't use SQL, but rather, use Dimensions and Metrics and the combinations of them.

Why is a reason for that? I think it doesn't have a SQL interface (or a plain web server log download)? If so, how do SQL statements translate to Dimension, Metrics (and Segment and Filters)?

It seems that Metrics tend to be the "aggregates" such as count() or average(), and Dimension tends to be the logged values themselves (such as Browser == IE or Country == Australia), which is the same as the group by values. Filters is like conditionals, and what about Segment?

It seems that if we specify Dimensions, then it automatically does a group by and display that field as well. It does count() or sum() usually. What if we want average(*) instead? And what if we want it to show but don't want it to do a group by?

example website to experiment is at http://code.google.com/apis/analytics/docs/gdata/gdataExplorer.html

Answer 1

use of the terms "Dimensions" and "Metrics" suggests that Google are using an OLAP database rather than a relational database.... SQL is used for relational databases: OLAP uses MDX or proprietary query languages (if Oracle).

From http://en.wikipedia.org/wiki/OLAP

The core of any OLAP system is an OLAP cube (also called a 'multidimensional cube' or a hypercube).

It consists of numeric facts called measures which are categorized by dimensions.