When loading a spreadsheet .xlsx file (either by uploading or by email), Scoop will analyze the contents of your spreadsheet attempt to extract tabular data.

Where does Scoop look for tabular data?

1. When you upload a spreadsheet, or setup an email inbox, or connect to a live Google sheet, you have the opportunity to specify a named range to utilize. This is the best practice
   1. Identify the table of data you are interested in in your spreadsheet
   2. Create a named range
   3. Tell Scoop that is the range to use for your dataset. Note that named ranges can come in a couple of different forms. For example you could name a range A1:E5 which is a discrete range, but you can also name a range A:E which means to load everything in columns A through E. In this way, as you continue to add data, the range includes it automatically. Alternatively, you can specify a range which goes beyond your data to accommodate any future rows.
2. If you do not have any Scoop ranges in your worksheet, Scoop will attempt to isolate tables from the first sheet in your spreadsheet workbook.

De-pivoting of Dates

A very common pattern in a spreadsheet is to have a series of attribute columns and then have a series of columns across the top that represent a single metric over many periods (say months). If those dates are proper dates (not date labels), Scoop will de-pivot those columns into a date attributes so that you can do analysis in Scoop over time.

In the example above, you can see this pattern. In the case where there is a single date series along the top, and the dates are actual dates (not text), Scoop will de-pivot those dates into rows which will allow you to analyze that data over time. By default, Scoop will create a new column with the name of your range as the metric.

And during upload:

Which then creates a Scoop dataset:

Allowing you to analyze your data flexibly: