As with all marketing, audience targeting is one of the major factors of ensuring that your creative is being seen by the correct prospective customer or lead. Direct Mail is one of the only offline channels that allows you to have detailed targeting down to the recipient-level rather than just the overarching household level.
Postie can use your CRM data to unlock various capabilities. The most common capability is creating a lookalike model. We feed your CRM data through the Postie machine learning stack and then through our DMP to locate and isolate customers and prospective customers that look and act just like yours.
The most common audiences to upload would be of best customers or highest LTV. The larger your dataset, the more precise our targeting can be. Our data science team recommends you upload a CRM audience of approximately 14,000 to 20,000 customers. This allows for a statistically significant model to be built.
Step 1: Log-in to the Postie platform
Step 2: Select “Audiences” in the left navigation
Step 3: Choose "New Audience" in the top right corner
Step 4: Choose "Upload an Audience File"
Step 5: Choose the audience file directly from your computer
Step 6: Name this file with the exact nomenclature that the rest of your business uses.
This is going to make it very easy to find, whether it is within your campaigns or you just need to find that audience to look at an audience portrait analysis.
You also have the ability to download a sample CSV to see the correct parameters that are necessary for uploading any audience file into the Postie platform.
Step 7: Click Create
Step 8: Map your columns to the respective areas within the platform.
This allows for seamless USPS and CASSing validation: first name to first name, last name to last name, address 1 to address 1, address 2 to address 2, so on and so forth.
If your first row of the file is a header row and should not be imported, please keep this box checked.
Step 9: If everything looks fine, click Import.
Your Postie Client Success Manager will now kick off the model build with our Data Science team. This process can take up to 48 hours.