· Saul Mateos · CFO Career · 7 min read
CFO Leadership Lessons: Feedback, Exits and Ownership
CFO leadership lessons from the seat: why soft feedback develops nothing, how to fire with dignity, and who gets to say stop on an open item.
The first of my CFO leadership lessons: softening feedback feels kind. It develops nothing.
CFO leadership lessons start with tough feedback
The flinch when someone hears the real thing isn’t damage. That flinch IS the development.
Feedback is painful. Most people don’t enjoy receiving it, and most managers don’t enjoy giving it. Doesn’t matter. It’s still the job.
And it hits harder in some cultures than others. I grew up in Spain. Receiving direct feedback took me years to stop hearing as an attack and start hearing as a gift.
The leaders who shaped me most gave me the hardest feedback. The kind that stung, then stuck. I didn’t enjoy a single one of those conversations. I’m grateful for every one.
Early in my career I learned the sandwich. Compliment, hard thing, compliment. It’s a real tool for low-stakes feedback, and it works. The problem was I kept using it when the stakes had outgrown it. I told myself I was being kind. I was being cowardly. The person stayed in the same patterns for another year. Then another. Eventually they hit a ceiling they couldn’t see coming, and I’d helped build it.
That’s not kindness. That’s negligence in slow motion.
The trade I didn’t understand at the time:
Soft feedback buys you comfort today. It costs them growth for a year.
The hard version is the one that lands. Specific over diplomatic. Direct over indirect. Graceful, never cruel. The kindest version of honest is still honest.
A few things I had to learn:
If they don’t push back, you didn’t say anything real. The discomfort you see across the table is them rewiring on the spot. That’s the part you came for.
If you’re delivering the same feedback for the third time, the problem isn’t that they didn’t hear it. The problem is they did, and nothing happened next.
If the conversation ends with both of you feeling good, you probably traded their development for your comfort.
And one rule I learned the hard way: never deliver feedback when you’re hot. That’s not feedback. That’s lashing out in feedback’s clothes. Wait for the heat to drop. Then say the same hard thing, cold.
A few months ago I wrote that the best thing you’ll build isn’t a process, it’s a person. Every 1:1 is a deposit. Every tough feedback conversation is the actual investment.
Every softened version is a withdrawal you’ll pay back later, usually at the worst possible moment.
If you can’t remember the last time someone you lead pushed back on your feedback, they’re not the problem.
You are.
The best thing you’ll build isn’t a process. It’s a person
As a leader, the best thing you’ll build isn’t a process. It’s a person.
I spent years perfecting systems. Forecasting models. Dashboards. Automations that took months to get right.
None of them survived.
The forecasting model got replaced when we switched ERPs. The dashboard died when we changed BI tools. The automations broke the moment we updated the tech stack. I’ve watched six figures of work get deprecated in a single quarter.
You know what didn’t get deprecated?
The analyst I trained six years ago is now a Director of FP&A at a Fortune 500. The coordinator I mentored runs finance at a $200M company. The intern I invested time in just got promoted to CFO.
They took what I taught them and built things I never would have.
Your spreadsheets have an expiration date. Your people don’t.
Every 1:1 is a deposit. That tough feedback conversation where they look at you like you’re crazy? That’s an investment. Every “here’s how I actually think about this” is a transfer of operating system.
The legacy that lasts isn’t the work. It’s the people who carry the work forward after you move on.
Firing with dignity is part of development
Early on, I held on to someone too long. Told myself I was being kind. I was being a coward.
The conversation scared me more than the cost did, so I let them sit in a seat they couldn’t grow into. That “kindness” spent a year of their life they could’ve spent finding the right one.
What took me too long to see: firing someone is part of how you develop them.
Most of the time the person isn’t bad, the person and the job don’t fit. Someone who ran one slice at a 30,000-person company can’t suddenly own the whole thing at a startup. Treating that like a character flaw is its own disrespect.
Every exit is a broadcast. The team reads it for one thing: do you get discarded here, or treated like an adult? Do it with dignity and the people who stay trust you more.
When it has to happen:
- Direct. Say the real thing once, cleanly.
- Fast. The day you know is the day they deserve to know.
- Never an ambush. A surprise means you failed months ago.
- They hear it from you. Not HR, not the grapevine.
Holding someone in the wrong seat isn’t the kind option. It’s the cruel one in a nicer coat.
The best thing you’ll ever build isn’t a process. It’s a person. That doesn’t stop being true on the way out.
Somebody has to be able to say stop
Three departments. Same open item. None of them could close it.
I’ve seen this in every operation I’ve run. The first team flags an issue and escalates it off their plate. It lands on the next desk. That desk can’t close it either, so they pass it to a third group. The third group sends it right back to the first team. Like a hamster on a wheel.
Three departments touched it. None of them owned it. And one open issue bouncing across three desks costs three people’s attention. That’s the hidden headcount tax nobody budgets for. The to-do list never shrinks. It just recycles.
Here’s what I keep writing down: every process needs a named person with finality authority. Not a better task tracker. Not a new template. A person who can look at an open item and say “stop.”
Most org charts name who can say “go.” Almost none name who can say “stop.”
The fix was right. The call wasn’t his to make
Someone on my team changed a process overnight. I found out because something else broke.
The problem he fixed was real. Several people were doing the same work on the same accounts without knowing it. So he rerouted everything through one person.
Reasonable fix. Except the old route had one step he didn’t know about, and the person who owns that process did. The new route skipped it. A batch of accounts sat untouched for a week before anyone caught it.
The conversation I had with him wasn’t about the fix. The fix was good.
It was about who he didn’t talk to. Before you change a process, talk to the person who owns it. Then talk to the people downstream of it. They know which step looks redundant and isn’t.
None of that runs through me. My job isn’t to sign off on every change. It’s to make that conversation the normal first step, so nobody here changes a process alone.
Solving a problem is a skill. Owning a change is a discipline. He had the first. Most of us learn the second one the hard way.
Good intent. Real fix. One conversation short.
The part of the job nobody warns you about
“Behind Blue Eyes” came on Spotify the other day. I turned it up without thinking.
Then it hit me. That song is the CFO job.
You carry the number nobody wants to hear. You name the constraint while everyone else is still celebrating the plan. You sit in the room knowing you’re about to be the least popular person in it. Not because you’re wrong. Because you’re the one who has to say it out loud.
Nobody warns you about this part. They talk about the models, the board decks, the strategic influence. Nobody mentions the night you’re alone with a forecast you wish looked different, knowing you still have to present it tomorrow.
The loneliest version of the job isn’t the hard calls. It’s the space between knowing and saying. That gap where you carry it alone because the org isn’t ready to hear it yet.
I stopped waiting for that gap to feel comfortable. It never does. The job is saying it anyway. And I still wouldn’t trade it for any other seat.