Chapter 9
At fifty-six, I made the mistake I had spent half my career warning other people about.
Not the same mistake.
Same shape.
Benton Group was opening a new hotel near O’Hare.
Opening date:
October 15.
Convention contract:
three days later.
Everything tight.
Fire inspection scheduled:
October 10.
One section of decorative wall paneling near a service corridor had not received final flame-spread documentation from the manufacturer.
Contractor said:
product matched approved sample.
Architect agreed visually.
Fire consultant said:
“Hold final acceptance until certificate arrives.”
Not:
unsafe.
Not:
remove.
Documentation hold.
The corridor was not open to guests yet.
Then supplier emailed:
certificate delayed because laboratory portal was down.
Expected:
forty-eight hours.
I said:
“Can we finish punch list and close the wall package pending the PDF?”
Project manager said:
“Consultant asked us not to close acceptance.”
I answered:
“I’m not asking you to mark accepted.”
“Just keep moving.”
There.
Keep moving.
Two words.
It sounded harmless.
Then contractor interpreted:
install remaining matching panels.
Because:
keep moving.
The next morning, fire consultant arrived.
Saw installation expanded.
Stopped.
“Who released this?”
Project manager said:
“Operations said keep moving.”
My stomach dropped.
I had not released.
I had still created the ambiguity.
The certificate arrived that afternoon.
The product complied.
No safety failure.
No rework needed.
The easy response would have been:
see, fine.
I knew better.
I opened an internal review.
My own email included.
The finding:
I had not formally overruled the consultant.
I had issued ambiguous schedule pressure while a documentation hold remained.
The project manager and contractor should have verified scope.
I should have said:
continue unrelated punch-list work; do not install or accept additional paneling until fire consultant releases.
Specific.
I received:
formal executive coaching,
and my performance review noted the incident.
Some board members said:
“Is that necessary? The product passed.”
Yes.
Because outcome does not rewrite instruction quality.
Then we changed:
all project holds had written scope and release authority.
Again.
Different company.
Same human problem.
I had spent years teaching it.
Still did it.
That was humbling.
Then one manager asked:
“Does this mean we can never say ‘keep moving’?”
No.
It means say:
what may move.
Language carries authority differently depending who speaks.
A junior coordinator saying:
keep moving
might be encouragement.
A vice president saying it during a hold can sound like release.
Power changes ordinary words.
Then I thought of Vanessa.
For the first time, I understood one part of her mistake from inside authority.
Not the grabbing.
Not the threat.
The schedule pressure.
You know what you mean.
You assume everyone else will preserve the boundaries in your head.
They cannot see them.
That insight did not excuse her.
It made me less simplistic.
I did not write to her.
Accountability does not require reunion.
I kept the lesson at work.
Then Adrian made his own governance error.
Romano Holdings invested in:
a restaurant technology company.
His nephew Anthony ran diligence.
Adrian liked the founder.
A lot.
He told Anthony:
“Don’t kill this over paperwork.”
Anthony later waived:
one minor insurance-document requirement
without formal exception approval.
Nothing bad happened.
During audit, the waiver surfaced.
Adrian asked:
“Why did you waive it?”
Anthony said:
“You told me not to kill the deal over paperwork.”
Adrian went silent.
Then laughed bitterly.
Of course.
He reported his own statement to:
governance committee.
The committee did not discipline him formally.
They did require:
executive deal comments documented as:
preference,
not approval.
And exceptions needed:
written owner.
Adrian later told Michael Trent:
“I became Vanessa for one sentence.”
Michael answered:
“No.”
Then:
“You made one similar authority mistake.”
“Do not turn accountability into melodrama either.”
Good.
People are not one mistake.
Patterns matter.
Then Matteo died.
Eighty-three.
Peacefully.
Adrian attended funeral.
No press invited.
Matteo’s will left Adrian:
one box of Carlo photographs,
and a letter.
The letter said:
I thought keeping Stefano hidden protected you.
It protected me from having to tell you something difficult.
Those are not the same.
Then:
I am sorry I confused them for so long.
Adrian gave the letter to R-17 archive after keeping a copy.
Not because he wanted public history.
Because future family trustees needed:
one example of how secrecy can become self-protection.
Then the Romano family trust adopted:
mandatory review of long-term sealed instructions.
Any instruction sealed more than five years required:
independent counsel review to determine whether conditions still made sense.
Not automatic disclosure.
Review.
The reason was Carlo’s instruction:
do not tell Adrian unless Stefano can participate
had remained frozen even as:
Adrian became an adult,
Carlo died,
Stefano’s situation changed,
financial interests were mishandled.
Dead instructions need stewardship too.
Then a family member objected:
“You’re violating Carlo’s wishes.”
Independent counsel answered:
“No. We’re checking what his wishes require now.”
Important.
A trust should not confuse literal silence with faithful administration.
Then at Benton, I began planning retirement.
Not yet.
Three years away.
I trained successors.
One woman:
Marisol Grant.
Brilliant.
Direct.
Former engineer turned hotel operator.
She challenged me constantly.
Good.
In one meeting, I said:
“We’ve always done it this way.”
She raised an eyebrow.
I laughed.
“Withdrawn.”
Then she proposed:
incident appeals for frontline staff.
If manager closed a report too quickly, employee could ask second review without going through same manager.
I loved.
I had not built it earlier because systems always have another blind spot.
Then first appeal:
a bellman reported a loose exterior handrail.
Manager closed:
maintenance ticket open.
Bellman appealed:
he thought area should be blocked until repair.
Second reviewer agreed.
Temporary barrier.
Repair same day.
No injury.
Then another appeal:
server reported ballroom floor “unsafe” because newly polished.
Second reviewer tested.
Within specification.
No closure.
Employee still protected for good-faith appeal.
A system should allow:
being wrong without being punished for asking.
Then I met Adrian at an industry dinner.
First time in four years.
He was older.
Gray in beard.
Still controlled.
Less interested in being impressive.
He said:
“Emily.”
“Mr. Romano.”
“Still?”
“Yes.”
Then:
“I heard about your fire-panel incident.”
I stared.
“Industry gossip?”
“Michael.”
“Of course.”
He smiled.
“Product passed.”
“Yes.”
“And you still wrote yourself up.”
“Yes.”
Then:
“Annoying, isn’t it?”
“What?”
“When the outcome is fine and the process was still bad.”
I laughed.
“Yes.”
Then he raised his glass.
“To paperwork.”
I raised mine.
“Don’t get sentimental.”
That was the first time our past felt like professional history instead of an open wound.
Before I retired, Marisol proposed an executive-request log.
I resisted.
“Too bureaucratic.”
She showed me six recent examples where senior leaders had asked for exceptions verbally.
One was mine.
A vendor had been allowed into a loading dock outside normal hours because I said:
“Make it work tonight.”
Security had interpreted that as permission to waive the standard badge process.
I meant:
find an approved after-hours path.
Again, the boundary in my head had not traveled with the words.
We created the log.
Not every executive comment.
Only requests involving:
access,
safety,
record exceptions,
spending authority,
or policy waivers.
The request could still be approved.
The point was to identify who actually owned the decision.
Within months, senior leaders changed how they spoke.
More:
“Can you route this?”
Less:
“Just do it.”
That was culture becoming visible in grammar.
The executive-request log created one uncomfortable entry involving Marisol too.
Before she became COO, she asked security to let a contractor stay after normal loading hours because a ballroom opening was behind schedule.
Security logged the request.
Facilities reviewed:
after-hours work was permitted,
but only with an additional fire watch.
Marisol had not known that condition.
The contractor stayed once the fire watch was assigned.
Later she told me:
“I thought the log was there to catch executives doing something wrong.”
I said:
“It is there so the person who knows the missing condition can see the request.”
That was a better description.
Governance is not a trap.
At its best, it is a way to connect partial knowledge before action becomes irreversible.
As retirement approached, the board asked me to list “key people the next COO should keep.”
I refused the wording.
I gave them:
roles,
performance history,
critical responsibilities,
and succession risk.
The reason was simple. “Keep this person” turns a predecessor’s preference into pressure on someone who has not yet taken the job.
Marisol needed room to evaluate her own team.
I could document what I knew.
I could not pre-own her decisions.
That lesson sounded far removed from R-17.
It was not.
So much of that old disaster began with one generation deciding what the next person should be allowed to know, choose, or control.
The board eventually changed the succession questionnaire itself.
Instead of asking outgoing executives:
“Who should the next leader trust?”
It asked:
“What information does the next leader need to evaluate this area independently?”
That was a much healthier inheritance.
Trust could be earned again.
May you like
The record only had to make the starting point visible.
One final succession rule came from that exercise: every inherited exception needed an owner and a review date. If nobody could explain why an old waiver still existed, the next leader was allowed to reconsider it instead of treating my past approval as permanent authority.
