At my son’s old school, they had a very nice lady who took care of attendance. If our son was going to be late or absent, we’d call or email her. It seemed simple enough.
Until one day when I emailed the school to report an absence and our son was marked with an unexcused absence anyway.
I was surprised and more than a little annoyed. Our school district only allowed so many unexcused absences before students could be declared truant. I went to find out what had happened.
Apparently, my email never made it to the school.
These things happen. In this case, the email had become stuck in a sending loop in my email platform. I could prove that I had attempted to send it, but the system wouldn’t budge. The absence remained unexcused.
At first, it felt like a simple technology problem.
But the more I thought about it, the more I wondered whether it was really an email problem at all.
The attendance process looked straightforward on the surface:
Parent reports absence → School records absence
But the actual process was much longer:
Parent sends email → Email provider works → Internet connection works → School email server works → Attendance person receives email → Attendance person is present → Attendance person processes email → Attendance system is updated
That’s a lot of assumptions hiding inside a process that everyone thinks is simple.
And this wasn’t the only time the system broke down.
There were days when the attendance person was absent. Attendance emails went directly to her personal work inbox rather than to a shared attendance account. No one else had access to her voicemail. No one else was monitoring her email.
The system worked beautifully right up until the moment the person carrying the system wasn’t there.
So our family created a workaround.
We called and sent an email.
The school told us we didn’t need to do both.
But experience had taught us otherwise.
What I find interesting isn’t who was at fault. My email failed. Technology sometimes fails. People get sick, like the attendance lady. People take vacation, which they should be able to do. None of those things are unusual. None of these things are unreasonable.
The interesting question is why the consequences of those failures fell entirely on the families using the system.
Because the burden didn’t disappear when the process broke down.
It found a home in the families.
When the attendance system couldn’t tolerate a missed email, the burden shifted to parents to verify delivery.
When the attendance person was absent and nobody could access her inbox, the burden shifted to parents to figure out who to contact.
When there was no mechanism for handling technical failures, the burden shifted to parents to prove they had done the right thing.
The organization still needed attendance information. The work still needed to happen. The risk still needed to be managed.
But instead of being carried by the system, it was carried by the families using it.
That’s why our family started calling and emailing.
Not because we enjoyed doing duplicate work.
Not because we didn’t trust the attendance clerk.
But because experience had taught us that the official process wasn’t always sufficient to protect us from the consequences of system failures.
The workaround became part of the process.
What’s remarkable is how often this happens.
When organizations cannot absorb variability, users absorb it instead.
Parents create backup procedures for school attendance.
Patients keep their own records in case they need to prove what happened.
Clinicians maintain shadow spreadsheets.
Administrators develop manual tracking systems.
Everyone quietly compensates for the weaknesses of the process until the compensation becomes normal.
While all of this shadow work is being done, leadership may believe the system is functioning well because the work is still getting done.
Reports are submitted.
Attendance is recorded.
Assessments are scored.
Data appears where it is supposed to appear.
On the surface, everything looks fine.
But the system isn’t succeeding because it was designed to succeed.
It’s succeeding because people are compensating for its weaknesses.
One of the most important questions in governance is not, “Who made the mistake?”
It’s, “Where did the burden go?”
Because every process failure transfers effort, risk, time, uncertainty, or accountability to someone.
The question is whether that transfer was intentional, visible, and fair.
A well-designed system expects emails to occasionally fail.
A well-designed system accounts for people being absent.
A well-designed system assumes that technology glitches and human absences are normal operating conditions, not extraordinary events.
The attendance clerk wasn’t the problem.
Looking back at the school attendance problem, the process was attached to a person rather than a function.
Imagine instead that attendance notifications went to a shared attendance inbox and voicemail.
The attendance clerk would normally monitor those communications.
If she were absent, another administrator could step in.
If that person were unavailable, responsibility could pass to a designated backup.
The specific titles matter less than the principle.
The information remains accessible.
Responsibility remains clear.
The process continues.
No individual has to be permanently available for the system to work.
I see versions of this pattern in healthcare organizations all the time.
Assessment alerts go to one administrator.
Outcome reports go to one clinician.
Vendor notifications go to one staff member.
Data exports are performed by one person who knows the exact sequence of clicks.
Everything works.
Until it doesn’t.
Then everyone discovers that the process wasn’t owned by the organization.
It was being carried by a human.
Good governance isn’t about making people work harder.
It’s about making sure the system can survive ordinary human realities.
Because good systems don’t assume people are permanent.
They assume people are human.
What Conditions Need to Exist for a Process to Reliably Work?
This attendance story may seem small, but it highlights several conditions that need to exist for any organizational process to function reliably.
Clear ownership.
Someone must be responsible for ensuring the process works, monitoring failures, and improving it when problems occur.
The function must be separable from the individual.
The process should continue when someone is sick, on vacation, changes roles, or leaves the organization.
Backup pathways must exist.
When one communication channel fails, another path should exist to prevent the entire process from stopping.
Critical information must be accessible to authorized staff.
Shared inboxes, documented procedures, and role-based access prevent knowledge from becoming trapped with a single person.
Exceptions must be anticipated.
Systems should assume that emails fail, technology breaks, and humans make mistakes. These are normal operating conditions, not rare events.
Accountability must be matched to control.
People should not face consequences for failures that occur entirely outside of their ability to influence or detect.
The process must be documented and visible.
If the process only exists in one person’s memory, the organization does not own the process. The individual does.
Data governance is the set of policies and procedures for handling data, including where data handoffs occur, who is responsible for the data (like the student attendance), and where the data are collected and stored.
When these conditions are present, organizations become more resilient.
When they are absent, work is often carried through heroics, workarounds, and institutional memory.
The goal of governance is not bureaucracy.
The goal is making sure important processes continue to work even when technology fails, people are absent, or circumstances change.
Governance begins when a process no longer depends on luck, memory, or a specific person to succeed.
