Let us say you have conducted interviews and had long discussions with many survey respondents. The entire conversation with multiple questions have been transcribed for each interview and fitted into an Excel file.

See Here for what an Explorer file should look like: http://support.gavagai.io/support/solutions/articles/4000089155-how-do-i-know-if-my-file-is-in-the-correct-format-

Yes, it is possible to analyze conversational transcripts as long as it is in text form. However, please keep in mind, The Explorer is built to analyze comments and find common Topics over the texts. If you merge all of a conversation or transcript into one cell per person you have surveyed (and conducted the interview with) then the common Topics found from The Explorer might not be insightful when analyzing the data set when it is in that merged form.

In the example below, if you analyze the first question column (What did you think of Z?) then 33% of the comments will be flagged by the Explorer mentioning the Topic: Book, that is, Respondent 2. If you look at the second image, with all of the interview questions in one cell for each respondent then you can see 100% of the comments will be categorized by The Explorer for the Topic: Book, that is, Respondent 1, 2, and 3.  Please keep in mind The Explorer finds Topics from analyzing one column for each project. So, in this case, we suggest making three projects and analyzing each column (or Question) in separate Projects. Of course, in the end, the analysis is up to you. For example, if you don't care how the Topics line up for certain questions, you only care about what Topics were spoken about from the entire interview in general then merging all the answers from each respondent can work.  

Note: You can have all of the questions split and still have the file uploaded as one Explorer project. See here for more: http://support.gavagai.io/en/support/solutions/articles/4000125731-if-a-data-set-has-more-than-one-open-ended-question-involved-what-should-i-do-