On November 3, I presented the Census Bureau’s operational plan for the 2020 Census – the blueprint by which we’ll conduct the next census – to Congress and the public. I was excited to tell the House Subcommittees on Government Operations and Information Technology about the depth and strength of our plan. Today, I want to go over some of the highlights from my testimony about the operational plan and our preparations for the 2020 Census.

In 2013, in response to funding constraints, we prioritized the 2020 Census research program. At that point we established the end of Fiscal Year 2015 as a key milestone for releasing the operational plan for the 2020 Census. I was pleased to inform the Subcommittees that we had met that goal, and that our plan is supported by solid research, including the 2014 and 2015 tests.

I was proud to report that in 2020 we will no longer use the paper-and-pencil processes that have characterized each census since 1970. The operational plan lays out four key areas of innovation that will ultimately deliver $5.2 billion in savings to the American taxpayers. The 2020 Census will be the most automated census ever, and we’re developing technologies and systems that will increase the efficiency of administering the once-a-decade headcount. We’re taking full advantage of the opportunity to innovate and using off-the-shelf technological advances from the last 10 years.

The operational plan details our research on the infrastructure that we need to take the census online. Through testing and development, we’ve developed prototype systems that incorporate mobile technology and optimal work assignments. A key component of these efforts is the Census Enterprise Data Collection and Processing (CEDCaP) initiative, a new agency-wide approach to survey and census data collection and processing. We’re simplifying and integrating – moving to a small suite of shared, reusable systems instead of creating duplicative systems for each survey and census. This new, sustainable approach will enable us to conduct a modernized 2020 Census.

I also told the Subcommittees about our plans to manage the risks associated with delivering the most automated census ever, including:

• Delivering the 2020 operational plan three years earlier than in the 2010 census cycle. Releasing our plans now gives us additional time to communicate our plans and decision making process to Congress and other stakeholders.
• Making innovative use of existing technology and software. -to-date tools – instead of inventing our own – we can make the best use of technological advances for an accurate and cost-effective census.
• Developed a working prototype for the census that we successfully tested in 2015. Based on this prototype, we’ve drawn up most of the census’ specifications; we’ll finish this work based on the results of tests in 2016 and 2017.
• Establishing milestones for an end-to-end test of our systems. In 2018, we’ll hold a critical test of all of the census’ major systems. This end-to-end test will represent the culmination of research and testing, as we implement the planned census operations in real-time.
• Minimizing risk by making timely decisions. The operational plan shares a number of decisions that we’ve already made in preparation for the 2018 end-to-end test, and lays out a timeline for the remaining decisions that we must make.
• Recognizing the need for timely decisions on which systems we need to build internally, and which we need to buy from external sources. We’ve engaged the Carnegie Mellon Software Engineering Institute to guide us in this area and to ensure that this critical decision making process is successful.

Based on my experience in overseeing the 2000 Census and in the private sector, I am confident that the plan outlined above will lead to a successful implementation of our automation and systems development for 2020. Right now, we are on schedule to deliver a census in 2020 that is both innovative and cost effective – but we’re also at a critical juncture. In order to execute a 2020 Census that reduces costs while maintaining quality, we must receive adequate funding for the entire lifecycle. By investing now, we can save more than $5 billion while ensuring we produce an accurate and cost-effective Census.

I must emphasize again how pleased I was to have the opportunity to update our oversight subcommittees in the House of Representatives, and to tell them that we’re on schedule to deliver a 2020 Census that is that is both innovative and cost effective. By taking a proactive approach in researching and testing modern, groundbreaking methods, we can make the 2020 Census the most cost-effective and automated Census ever.