based on the raw data you already have, it would be nice to build a simple regression model. That way one could combine these factors in ones own case.

e.g. what is the expected average for a senior DS, in San Francisco, working at Google? you cannot just 'add' up the different numbers to understand this.

You could just publish it as a linear equation then maybe provide a small table of example values.

just a thought. seems pretty easy given your dataset?