You can write job results directly to your Twitter Tailored Audience.
- Basic knowledge of Arm Treasure Data, including the toolbelt.
- A Twitter Ads Account or an account with Ads Manager role.
- Authorized Treasure Data Twitter app access to your Twitter Ads Account
Create a new connection
Go to Treasure Data Connections. Search and select Twitter Tailored Audience.
The following dialog opens.
Select an existing OAuth connection for Twitter, or click the link under OAuth connection to create a new one.
Create a new OAuth connection
Log into your Twitter account in the browser (if you haven’t logged in previously) and grant access:
Or grant access to the Treasure Data app when you are already logged in.
You will be redirected back to Treasure Data Connections. Repeat the first step (
Create a new connection) and choose your new OAuth connection.
You can give a name to your newly created connection and save it for later use.
Configure to output results to Twitter Tailored Audience connection
Output results at the top of your query editor and select your Twitter Tailored Audience connection as follows:
There are several parameters to edit:
- Twitter Ad Account ID (required): This is your Twitter Ad Account ID. See Appendix C for getting account id.
- Tailored Audience Name (required): Name of Tailored Audience list to create.
- Important note: If you have many Tailored Audience lists with similar names as this input, the latest Tailored Audience Name will be used. We recommend that you give your Tailored Audience list a unique name.
- Audience type (require): Type of entry in the list. It is one of email, id, device id or handle.
- No need to normalize and hash records (default
false): Indicates whether the data has already been normalized and hashed. If not, TD automatically normalizes and hashes the records.
- Retry limit (optional, default
6): Number of retries before it gives up.
- Initial intervals in milliseconds between retries (optional, default
10000): Interval to retry if a recoverable error happens (in millisecond).
- Maximum intervals in milliseconds between retries (optional, default to
32000) : Maximum time in milliseconds between retrying attempts.
- Skip on Invalid Records? Ignore invalid records/errors and continue loading other records
Here is a sample configuration:
Write the Query to populate a tailored audience list
We support the creation of a new Tailored Audience list.
On the Treasure Data console, run the following query with
Output results into a connection of Twitter Tailored Audience (see Appendix A for more details of column naming):
SELECT email FROM ( VALUES ('firstname.lastname@example.org'), ('email@example.com'), ('firstname.lastname@example.org') ) tbl (email)
The preceding query will not match any real users as it’s solely for demo purposes only. Also, it requires no source table (for the ease of testing out this feature), but you still need to choose your database, so pick “sample_datasets” or any other arbitrary table.
The query should complete in a few seconds. Check the Twitter Audience Manager for your new list:
Twitter takes up to 24 hours to validate your newly created list. Therefore, the list will show the status as “Processing”.
Status changes to “Ready” (or an error if the list is not valid) within the next 6 to 8 hours.
Appendix A: Tailored Audience data
The Output result needs to follow the Twitter Tailored List data specifications here. We support single column, single type.
- TWITTER_ID: Twitter user IDs
- DEVICE_ID: IDFA, AdID or Android ID of Twitter users
- EMAIL: Email addresses
- TWITTER_HANDLE: Twitter users name aka. Screen names
Appendix B: Data Normalization and Hashing
Our result output can normalize your values automatically to follow Twitter’s normalizing rules. See here.
All values, uploaded to Twitter for matching, need to be normalized with the normalizing rules of Twitter. The values will narrowly miss chances to match if they are not normalized. If you need to normalize values by yourselves, apply your own normalization before submitting the output.
The conversion below is actually applied per type in our result output for normalization.
- Email Addresses : Lowercase, remove leading and trailing spaces; ex: email@example.com
- Twitter User Name : No @, lowercased and leading and trailing spaces trimmed; ex: jack
- Twitter User IDs : Standard integer; ex: 143567
- Device IDs : Lower-cased with dashes (IDFA), original format on device, not capitalized with dashes (AdID) and original format on device, not capitalized without dashes or spaces (Android ID); ex: 4b61639e-47cc-4056-a16a-c8217e02946 and af3802a465767e36
After normalization, the data must be hashed using
SHA256, without a salt. Additionally, the final output hash must be in lower case. E.g., 49e0be2aeccfb51a8dee4c945c8a70a9ac500cf6f5cb08112575f74db9b1470d and not 49E0BE2AECCFB51A8DEE4C945C8A70A9AC500CF6F5CB08112575F74DB9B1470D
Appendix C: Twitter Ads Account info
Access to an Ads Account is the pre-requisite to create the Tailored Audience List. The account ID can be retrieved from the Account Settings Page