We recommend you import data in XLSX (Excel) format but also currently support CSV.
Here is an example of what a typical file should look like:
Rules for formatting your file:
- 1 column per field and 1 row per record. In Ipiphany a field becomes an attribute of a record that you can analyze and/or filter by.
- The first row in the file should contain the names of the fields (columns).
- Each column should only contain data of one type of field. For example, a score column should only contain numbers.
- If a record does not contain a value for a particular field it should be left blank rather than being populated with 'NULL' or '0'.
- Date type fields should be an Excel compatible date.
- Verbatim Text type fields should be formatted as written. Please do not remove punctuation marks as these are used in the text analysis.
- The file should not contain more than 100,000 rows. If you need to import more row than this please import them separately by using the append feature.
Ipiphany works best when analyzing datasets containing both structured fields (scores, categories, dates, etc) and unstructured fields (verbatim text).
If Ipiphany detects records that have incorrectly formatted fields those records will not be processed and your dataset will display an orange error symbol:
Click this symbol to go to Settings > Data where you can download the failed records. Simply fix them up and append to complete your import.