Update tags via CSV

Upload a CSV file to set the values of several Tags for many devices at once, for example to prepare data for segmentation.

You can only update Tags that already exist in your account. The import doesn’t create new ones.

Youtube Video: Assign Tags to your subscribers' devices manually by uploading a CSV file

1. Prepare CSV

Export a segment and edit it, or create the file yourself.

Export segment

Go to Segments, select the Segment you’d like to set tags for and export the segment. The file contains:

• HWIDs of devices complying with that Filter segment (in the first column of the .csv)
• Push Tokens of user’s devices (in the second column)
• types of devices (integer and humanized)
• all Tags from your Pushwoosh account (in the remaining columns)

Change Tag values or add missing ones in the file, then save it.

Create CSV

To create the file yourself, set up these columns:

• Identifier column: name it HWID and fill it with device hardware IDs, or name it User ID and fill it with user IDs.
• Tag columns: name each one exactly as a Tag in your Pushwoosh account. Each cell holds the value to set for that device or user.

HWID	Name	Age	Push Alerts Enabled	SDK Version	First Install	Preferences
XXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX	Paul	25	false	6.4.12.0	2023-06-27 12:00:00	sport, news
YYYY-YYYY-YYYY-YYYY-YYYYYYYYYYYY	Anne	30	true	6.7.17.0	2025-11-14 16:00:00	fashion, travel

Representation of import.csv in plain text format:

HWID,Name,Age,Push Alerts Enabled,SDK Version,First Install,Preferences
XXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX,Paul,25,false,6.4.12.0,2023-06-27 12:00:00,"sport, news"
YYYY-YYYY-YYYY-YYYY-YYYYYYYYYYYY,Anne,30,true,6.7.17.0,2025-11-14 16:00:00,"fashion, travel"

List Tag values

A List Tag cell accepts two formats:

• Comma-separated list: sport, news, as in the example above. This format is easier to produce in spreadsheet apps. Values are always split on the comma.
• JSON array: ["sport","news"]. Use it when a single value contains a comma, for example ["Sports, general","news"].

Leaving a List Tag’s cell empty keeps the subscriber’s existing value for that Tag unchanged. To clear the subscriber’s value for a List Tag, use an empty JSON array ([]) instead. For Tags of other types (String, Integer, Date), an empty cell is imported as-is and overwrites the subscriber’s existing value.

2. Upload CSV

Caution

Before you upload, make sure that:

1. The file is saved in UTF-8 encoding.
2. Values are separated by a comma.
3. Every row has the same number of columns as the header, and every quoted value is closed.

A single malformed row gets the whole file rejected, not just that row.

1. Select the project. Tag values are updated in this project only. If an identifier isn’t registered in this project, its row gets no Tag values.

2. Go to Import CSV → Update Tag values.

   

3. Click Upload CSV and select your file.

   

Note

Columns that aren’t named exactly as Tags in your account are skipped.

Pushwoosh checks the whole file for parse errors when you upload it. A row that can’t be parsed (its column count doesn’t match the header, or a quoted value isn’t closed) gets the file rejected right away. A toast names the row and the reason, for example “Row 42: the number of values does not match the header.” Nothing is imported in this case. Fix that row and upload the file again.

3. Choose Tags to update

1. Choose the identifier type your CSV uses, HWID or User ID, and the column that contains it. The type decides which Tags can be updated:

   • HWID: device-specific Tags, such as SDK Version.
   • User ID: user-specific Tags, whose values belong to the user profile rather than a device.

   Pushwoosh shows which Tags will be imported and which will be skipped for the type you picked.

2. Select the columns to import.

3. Click Start import.

4. Check the results

The import status updates in real time. Large files can take a while, so you can leave the page and check back later.

When the import completes, the report confirms that tag values from the selected columns were updated. Check its counts:

• Matched: CSV rows, not devices, whose identifier found at least one device.
• Updated: devices, not rows. In a User ID import, one row can match several of the user’s devices, so this number can be higher than Matched.
• Updated users: in a User ID import only, the number of distinct users with at least one updated device. Two rows for the same user count as one.

Note

For a file over 100,000 rows, Pushwoosh processes it in batches and can count the same user as updated in more than one batch if that user’s rows land on both sides of a batch boundary. The Updated users count can then be slightly higher than the true number of distinct users.

A row is skipped for one of two reasons:

• No device matches the identifier. Pushwoosh checks it against HWID, User ID, or push token, regardless of which identifier type you chose in step 3.
• The identifier column is empty. The row never reaches device matching.

Matched rows plus skipped rows should add up to the total rows in the file.

Caution

In rare cases, the import fails after it starts, even though Pushwoosh checked the file at upload. This happens if the file changes between upload and Start import and a row can’t be parsed (its column count doesn’t match the header, or a quoted value isn’t closed). The status then shows Failed instead of Completed. The report names the row, for example “Row 42: the number of values does not match the header.” No tag values are applied, including for rows before the broken one. Fix that row and upload the whole file again.