Before launch, the platform will scan the validity of your address data (based on USPS flags) and return the number of mailable addresses, along with a CSV file listing all the invalid addresses. Common reasons an address may be labeled invalid include:
- The address has been flagged as vacant or inactive by USPS
- The address has an invalid primary number
- The address has missing or invalid secondary information (ie: apt. or suite #)
- The city/state/postal code combination is invalid
- The address belongs to a commercial mail receiving agency
After uploading your Audience or Mail File, Poplar performs a set of validations to ensure complete and deliverable addresses. The platform will assess the Total Records which is the number of correctly formatting fields in the file, as well as the number of Mailable Records which is the quantity that can be mailed based on your address strictness.
Click Download File with Error Report for a record of which addresses are invalid, and the reason they were marked as such (for example: "missing secondary data" indicates address_2 APT/Suite number is required to complete delivery). You can decide whether to address the errors or skip those addresses and proceed with the send:
Address Strictness can be set on an account-wide level from your Account Settings and a campaign level when creating a new campaign. This setting controls the level of address validation required to mail. To assure the highest likelihood of deliverability to we recommend the Strict setting. When mailing to commercial addresses, you may want to try adjusting to Normal or Relaxed depending on the confidence in your data.
||Only mails to addresses deemed "in-service" by USPS. New construction may remain "inactive" for a few months after being occupied.|
|Normal||Checks the existence of an address but ignore other data such as the USPS "in-service" flag.|
|Relaxed||Skips most address validation checks and should only be used for thoroughly vetted data or transactional mailings, where there is a high degree of confidence in the dataset.|