New Sawtooth customer. I inherited a fairly complex set of Sawtooth survey data from a consultant, and built some dashboards in Power BI. The consultant used Excel for their report, but we wanted to enable dynamic exploration of the data and relationships between responses.
Our survey had around 100 questions, with a variety of types: single choice, multiple choice, grid questions, ranked choice. I had the original Sawtooth CSV to work with, and unpivoted the data (We had around 12k responses, so this generated > 3m rows of data. )
I then had to generate a unique index for every question/row/column. Because each question type involved different numbers of columns and different column header formats in the raw data, I had to break out each question type and used Power Query to create an index based on that type. I created a control file for my model which allowed me to specify question type for processing (which btw I also used for sort order on the dashboard). Once this was done, I also used selected pivoted columns in the original dataset as slicers in a joined table.
I basically had to figure this out from scratch using our consultant's Excel reports to validate. I'm wondering if there are others out there who have done similar work, and if there are any shortcuts or best practices available? Have found only a limited number of references on this forum.