Know-How Exchange : Research/Metrics

Without knowing anything about how the questions were worded or how the sample was gathered or what made a valid respondent in each of the studies, I would think that with proper weighting that the two indices could be mathematically linked to provide an overall satisfaction index.
However, I would not rely on taking a simple average of the two scores to create an overall score.

You cannot take an average of the two scores, simply because they do not represent the same sample of respondents. There need to be separate surveys to gauge the satisfaction levels of respondents with respect to some pre-defined criteria- like range of products or shopping experience or ambience...

Satisfaction is a vague term if not supported by any contextual reference.

So, firstly define the context and then start measuring the indicive strength.

All the Best!!