organising_data/csv/Managing-Key-values-with-csv-files.md

 A spreadsheet is an easy way to keep track of the experimental variables associated with your data.  This examples goes through how to:
 
 1. Export metadata from a dataset to a csv file
+2. How to edit the key-value metadata in the csv file
 2. Add metadata to data set from a csv file
 
 In this example we will be using the guess-who images which can be downloaded from **add repo location**. 
@@ -24,7 +25,27 @@ The dataset now has an attachment (you may need to press the `Refresh` button ![
 
 
 ## 2. Edit the csv in Excel
-After downloading open it in Excel.  The file will have a single column 
+**1. Click the filename** to download it to your `Downloads` folder.
+
+**2. Open the file in Excel**.  The file will have a single column with `filename` at the top and the names of the images below it.  If this is a new dataset the rest of the spreadsheet will be empty.
+
+**3. Add the Key-Value metadata** Keys go across the page in the first column.  Fill in the values for each filename in the corresponding row and column below. Leave the cell empty if the image doesn't have a key-value associated with it.
+
+![Adding key-value metadata](./media/csv_in_excel2_50.png)
+
+**4. Save the spreadsheet as .csv**. Excel will warn you about loosing features, this is alright.
+
+## 3. Adding Key-Value pairs from the csv file
+
+**1.**  Select the Dataset **
+
+**2.**  Under `Attachments` click the `Attach a file` button ![](./media/add_icon_50.png)   
+
+![](./media/add_attachment1_50.png)
+
+**3. 
+
+