Secretary of Commerce nominee Wilbur Ross, who as a business school student served the Census Bureau as an enumerator in Boston, is scheduled to see a confirmation vote next week. If it is a “yes,” he should get to work immediately, looking to the report published this month by the Government Accountability Office that includes the 2020 Census among its list of “high risk” operations.

According to the GAO, the Census Bureau has announced a plan to replace its paper-and-pencil operation with smart technological innovations, but has yet to guarantee it can implement those changes successfully in 2020. The bureau scrapped some important 2017 field tests that could have helped answer that question because of uncertain funding. It also struggled with high nonresponse rates in 2016 tests, and cost estimates for the 2020 survey are shaky. All of that needs to change before it is too late for fixes.

There is a reason the census is part of the Constitution: The national head count is critical to a functional democracy. Not only is it a vital research tool, but the census also determines how many seats each state gets in the House of Representatives and how district lines are drawn. Any doubt about the data’s validity could cause a crisis in the 2020 redistricting battle – which is why it is essential that the bureau gets it right.

That has become tougher (and more expensive) in a society where living situations are getting more complex and heads harder to count. A steadier funding stream would help the Census Bureau conduct the tests it needs to make sure its new strategies work in 2020, but the GAO says the bureau will also have to strengthen its internal controls and communication.

That the 2020 report could lack integrity would be worrying on its own, and it would be even more so under an administration that has repeatedly displayed a disdain for data. Saving the census would give Ross an early opportunity to prove that he, at least, cares about accurate numbers.