Survey data can be very useful in practice, especially when appropriate correlating variables are used at the same time. For example: many corporations track and report on customer satisfaction data regularly in order to identify market trends. That is a useful endeavor, but the degree of data utility can be massively enhanced when other data is recorded with the survey response (e.g., things like age, gender, socio-economic status, past purchase behavior).
Qualtrics now includes a web service feature which allows for seamless integration of non-survey-specific data. This opens up huge avenues in data analysis, where covariates can be used dynamically in analysis of variance-type applications. Data can be integrated in XML, Query String, or JSON formats to fit whatever datasource is available to the researcher. In practice, each of the values retrieved from the web service are recorded as embedded data elements in the survey response. The Qualtrics system doesn't limit the number of values that can be stored as embedded data, so the possibilities with this are endless. Also, as the web service response is stored as embedded data, data fields can be piped into question choices, survey text, and can even be used in branching logic in the survey flow.