Written By: Greg McKay, Government & Public Sector Executive Leader, EY US
I will never forget standing before an angry legislative oversight committee shortly after being appointed to lead Arizona’s child welfare agency. The state was in crisis. The agency was viewed as irreparably broken.
Our child abuse hotline dropped nearly one-third of incoming calls. Backlogged investigations approached 19,000, ironically about the same number of children in foster care. Children waited endlessly for permanency and slept in government offices. We lost nearly 50 percent of our 3,000-person workforce every year. Tragic cases dominated headlines and public trust had collapsed.
Everyone in the room wanted answers. Especially me.
The legislature directed us to commission an independent assessment. Prestigious consulting firms and respected academic institutions eagerly pursued the work. Months later, they returned with polished reports, hundreds of pages long, complete with findings, recommendations, benchmark comparisons and exhaustive literature reviews.
There was only one problem. Nearly all of their evidence was already years old.
The recommendations were built on historical data, prior audits, research studies and observations that reflected conditions from two, three, and sometimes four years earlier. By the time those reports landed on my desk, the children, families, workforce, provider community, legal system, and operating environment had already changed. Yet there was enormous pressure to implement those recommendations as though they represented current reality.
That experience fundamentally changed how I think about government improvement.
Imagine asking an oncologist to treat cancer using a PET scan from two years ago. No one would accept that in any other profession. Yet in human services, we routinely ask leaders to manage today’s crises using yesterday’s information.
The reports were not wrong. Many contained thoughtful observations and sound recommendations. They were simply answering a different question. They explained why the system had struggled. They could not tell us what required attention today.
Eventually, the conversation shifted to technology. The prevailing belief became that Arizona’s child welfare challenges were primarily the result of an aging Statewide Automated Child Welfare Information System (SACWIS). The answer, we were told, was to implement a modern Comprehensive Child Welfare Information System, or CCWIS.
Technology matters. Modern systems are essential. But technology was not why children slept in offices. Technology was not why half our workforce left every year. Technology was not why investigations went unassigned or why permanency took years instead of months.
Those were management failures, not software failures.
So we pursued CCWIS modernization while simultaneously reforming how the agency operated. And before our new CCWIS ever went live, Arizona transformed from one of the nation’s lowest-performing child welfare systems into one of its strongest.
We did not achieve that transformation because we replaced a technology platform. We achieved it because we changed how we managed the work.
We built mechanisms that made performance visible every day. We empowered supervisors to identify and solve problems immediately. We created tiered accountability around operational results rather than historical reports. Most importantly, we shortened the time between recognizing a problem and taking action.
The information already existed. What was missing was the ability to access it and react. That lesson has stayed with me ever since.
For decades, government has invested heavily in systems designed to document work, support compliance and produce reports. Those investments were necessary. But documenting yesterday’s work is not the same as improving today’s outcomes.
The next generation of modernization must move beyond recording events and toward orchestrating action. Because child welfare doesn’t have a data problem. It has a reaction-time problem.
The agencies that will define the next decade will not necessarily be the ones with the largest data warehouses, the most sophisticated dashboards, or the newest case management systems.
They will be the ones that recognize operational signals as they emerge, understand which conditions require intervention, and respond while there is still time to change the outcome.
Children cannot wait for the next quarterly report. Children need systems that act before today’s operational problem becomes tomorrow’s tragedy.
Yesterday’s data won’t save today’s child. But timely action just might.
About the Author
Greg McKay
Government & Public Sector Executive Leader, EY US
Greg helps agencies deliver optimal outcomes for vulnerable citizens through technology transformation. Previously, he led Arizona’s Department of Child Safety, taking it from America’s worst- to first-ranked child welfare system. He also implemented the first end-to-end Comprehensive Child Welfare Information System in America.
The views reflected in this article are the views of the author and do not necessarily reflect the views of Ernst & Young LLP or other members of the global EY organization.
This post was contributed by an APHSA Strategic Industry Partner. The perspectives and opinions expressed are those of the author(s) and do not necessarily represent the views of APHSA.