This analysis report presents the results of the Informed Delivery® pilot conducted in partnership with the United States Postal Service (USPS) during the 2018 End-to-End Census Test. The scope of the Informed Delivery® pilot was to leverage the USPS’ Informed Delivery® technology to increase internet self-response across the first five mailings of the 2018 End-to-End Census Test contact strategy and the follow-up Nonresponse Followup operation (NRFU) mailing. The universe of Informed Delivery® users within Providence County, RI, was 12,198 at the beginning of the pilot and reached 16,191 by the conclusion.

Informed Delivery® is a free service provided by the USPS, and did not require the creation of an Interagency Agreement. The total cost to the Census Bureau for use of the technology was $0.

The Census Bureau implemented use of the Informed Delivery® technology through thirty nine distinct groupings or batches of Census mail pieces, called campaigns, that were executed and studied as part of the pilot. When an enrolled user was sent a mail piece from the Census Bureau, they received an email with a scanned image of that mail piece, as well as a hyperlinked image created by the Census Bureau. This ride-along image linked to the Internet Self-Response (ISR) questionnaire for the 2018 End-to-End Census Test. The intent of this pilot was to examine the viability of using Informed Delivery® to encourage residents to complete an ISR questionnaire.

Many metrics were collected as a part of this effort, including the number of emails sent, the number of emails opened, the number of links to the ISR questionnaire clicked, and the response rates of users once on the ISR questionnaire was started. This information was collected from five different mailings, which were aligned to four overarching cohorts as defined by the contact strategy:

1. Internet First English Only
2. Internet First English/Spanish
3. Internet Choice English Only
4. Internet Choice English/Spanish

Over the course of the analysis, the team looked at how response rates changed with each mailing and what effect the four cohort listed above had on those rates. The average open rate for an Informed Delivery® email was nearly 70 percent and stayed relatively constant through all mailings. The rate at which users would click on the ride-along image was greatest at 5.9 percent for all opened emails in the first mailing. The clickthrough rate steadily declined to a terminal value of 2.1 percent by the fifth mailing.

There was over a 50 percent chance of an individual completing a questionnaire that he or she started. This is strong evidence that creating a more appealing image and increasing marketing around the Census Bureau’s use of Informed Delivery® can lead to increased response.

In order to document and make adjustments as needed, lessons learned were captured throughout the pilot. The Informed Delivery® tool is relatively easy to use, but more planning needs to be conducted ahead of a full decennial census in order to properly organize and document each campaign. The methods for tracking in this document worked on a small scale but would need rework if the Census Bureau wanted to scale this program nationally.

There are many things the Census Bureau can improve upon in the future to take full advantage of the breadth of Informed Delivery® services. Working with USPS subject matter experts, creating interactive images, and generating more targeted messaging based on cohort, group, and mailing, the Census Bureau can improve the overall response rate by using Informed Delivery®.