About different update methods in Custom Data Import

Prev Next

When you send in new files for the Custom Data Import, you want to have your data updated in the correct way. Based on the data your file has, a different type of update might be needed.

We've built the Custom Data Import tool to suit different kinds of data needs. Depending on whether your data is marketing performance data, a list of orders, or a customer database, a different type of update method might work best.

You need to select the update method when configuring Custom Data Import.

Append and deduplicate

Appending and deduplicating only adds or changes data that has changed since the last update. It uses selected fields to look for existing data.

When to use

  • When your historical data might change, and your data uploads might include updated values for already imported data

    • For example, if you are every day uploading “last 7 days” of data. This means that you will have 1 day of new data and 6 days of data that was already imported. In this case, you want those 6 days to be updated and not just added again.

    • Another example would be, if you have a 14-day attribution window in your advertising data, you can send us a file with the last 14 days of data every day, and we update the data for the days included in your file, but days before will be intact and will remain available for you to query.

    • Or, if you want to send a file with a list of orders, you can send a daily file with orders that have been updated in the past day and we will add new order IDs to your dataset and update the order IDs that you have already sent us.

Append only

With append, all new data you send is added to our database without updating or removing existing data. It keeps all data, but may cause duplicate date rows.

When to use

  • When data uploads don't include any overlapping data

  • When historical data won't change

  • When you’re sending only one day of data at a time

Full refresh

Full refresh completely replaces the existing data with new data each time a file arrives.

When to use

  • When you know that the data you have sent earlier can still change

  • When you want to have only a current snapshot of the data

  • When you’re sending a full dataset with every upload