How to Fire Someone

If you’re a finance leader in the SF area, I’d love to buy you a Miller Lite or three on Thursday February 12th. I’m hosting it at an undisclosed Bay Area location, guarded by a squad of ferocious miniature Bernedoodles.

Seriously though, let’s all have a fun night of “networking” (ugh I hate that word).

RSVP below to meet fellow nerds.

This week we have the following CFOs answering your reader questions:

They are two gentlemen I’ve looked up to for a long time, and I couldn’t be more pumped to get their input on the following topics:

Let’s get into it!

John from Calendly:

This is actually a great first assignment for your new controller. I’d ask them to own the transition end-to-end and come back with a clear project plan. In parallel, lean on the fractional firm — they should be very used to this. The big things to confirm are:

If things are in decent shape, a clean transition usually takes 30–60 days. The goal isn’t perfection on day one — it’s continuity, audit readiness, and no loss of institutional knowledge.

Steve from Bloomerang:

Non CPA here.

Congrats on hiring a Controller. The hard part is already done. Seriously. Picking the right full-time human is way tougher than sunsetting the fractional firm that’s been maintaining your books for the last few years. Now it's just project management.

Your goal should be to wrap the transition in 30 days. No reason it should take longer, unless they’re holding your chart of accounts hostage. Start by asking for a clear list of what they currently own. Close process, reconciliations, payroll, rev rec, state tax filings. If they can’t give you a checklist, even more reason to move on.

Have your Controller shadow a full close cycle from start to finish. They should walk out with the keys in hand and confidence to drive. Ask the fractional team what’s “weird.” Journal entries, manual workarounds, audit adjustments. That’s the stuff that will bite you in three months. Also how quickly your Controller wants to take everything is a great first test.

John from Calendly:

You’re ultimately going to be working with a person, not a logo — so partner quality matters a lot. But it’s not a 50/50 tradeoff.

I’d work backwards from talent and distribution. In competitive markets, especially in software, your ability to attract great people is one of the few durable advantages you can create. Like it or not, employees and customers often outsource their own judgment to your investors. A strong firm brand still carries real signaling power.

My bias: don’t ignore brand lightly. If you love the partner at a newer firm, make sure you’re clear-eyed about what you’re giving up — network, recruiting pull, credibility in future rounds. The right answer isn’t universal, but the costs are real.

Steve from Bloomerang:

Trust your gut. You don’t work with a logo, you work with a person. A big name on the cap table might not return your calls when you miss plan. A great partner at a lesser-known fund will roll up their sleeves and help you figure it out.

Sure, a name-brand firm can help with recruiting, signaling, and future fundraising, but only if the partner doing your deal has juice and shows up when it counts. A logo can’t jump on a Sunday call. A great partner can. And will.

I’m a Midwest guy. We tend to pick people we trust over people who own Patagonia vests with their fund’s logo on them (kidding I love a good vest). So my rule of thumb? It’s 80% person, 20% firm. Go with the investor who’s going to get in the trenches with you, not just pose for the Series A press photo. This is a marriage. Pick the person who’s going to show up when it’s hard, not just when it’s headline-worthy.

responses…

John from Calendly:

The most important part of a performance-based termination is everything that happens before the final conversation. If you’ve done your job well, the person should never be surprised. Use every mechanism you have — 1:1s, written feedback, performance reviews — to make expectations explicit.

One tactic I rely on is asking the employee to summarize my feedback back to me in writing. If they can’t clearly articulate the gap and what needs to change, that’s a signal you’re not aligned yet.

By the time you reach the termination conversation, it should be short and direct. Something like:

“Today is your last day with the company. We’ve had multiple conversations about performance, and we haven’t seen the change we needed. Thank you for your contributions, our People team will walk you through next steps.”

At that point, the decision is made. Clarity is kinder than debate.

Steve from Bloomerang:

Firing someone sucks. Anyone who says otherwise is a sociopath.

First rule: this is not a debate. You’re not asking for agreement. You’re delivering a decision. Be direct, firm and kind, in that order.

Here’s a structure that works:

1. State the decision clearly.

“I want to let you know we’ve made the decision to end your employment with us, effective today.”

2. Don’t over-explain. You’ll feel tempted to fill the silence. (This has been my biggest struggle personally) Don’t. Rambling invites pushback.

“This decision is based on overall performance and fit for the role.”

3. Move to next steps.

“Our Head of HR will walk you through logistics like final pay and benefits.”

If they push back, calmly restate:

“I understand this may be difficult to hear, but the decision is final.”

Then stop talking.

You can still be human.

“I appreciate the work you’ve done here and wish you the best going forward.”

It’s not easy. But avoiding it makes it worse. Be clear. Be respectful. Be done.

John from Calendly:

I’d treat your board deck like a product and your board like customers. Start by talking to them directly: what do they actually want to see, what’s been most useful in other boardrooms, and what do they find distracting or redundant? That’s not a sign of weakness — it’s a founder showing good judgment and empathy.

From there, be ruthless about your word and slide count. Boards care about a small number of leading indicators tied to your strategy, not a perfect representation of the business. If you’re operating in a fast, experimental environment, name that explicitly and focus on:

One practical tip: I built a no nonsense board member GPT and run all board content through it. Today, I do not publish anything to the board without the bot’s advice.   

Steve from Bloomerang:

First board meeting? Deep breath. Now repeat after me: governance is not the same as operating. As my CEO says, ‘the board's job is not to check your homework.’

The biggest trap I see, and still fall into myself, is trying to compress all your operational context into a deck and expect the board to absorb it. They won’t. They shouldn’t. That’s not their job.

Your job is to synthesize, not to download.

Pick ~5 metrics that reflect the health, efficiency, and velocity of the business. If you’re in a flat or declining efficiency (FDE) environment, that probably means Rule of 40, CAC payback, burn multiple, bookings efficiency, and employee productivity. Frame them in terms of tradeoffs: what you’re investing in, what you’re pulling back on, and what the return looks like.

Remember, the board doesn’t need to see every page of your notebook. They need to understand where to focus and where to push. Give them the headline, then let them ask for depth. And candidly, I’m still working on this. The instinct to show how much you know is real. But the board doesn’t need your fidelity. They need your judgment.

Less data, more signal. You got this.

John from Calendly:

I’d avoid starting from “everyone else wants one” — that’s not a decision framework, it’s a social dynamic. Your execs ought to know better.

First, look at the health of the business. The more constrained you are, the higher the bar should be.

Second, get specific about the problem: what work is not getting done, or is being done poorly? For example, is the request because of time, context switching, or travel load?

Third, exhaust software first. Scheduling, travel, and inbox triage are increasingly solvable with good tools.

That said, software replaces tasks; great EAs increase leverage. If the need is real, I usually start with shared support. For example: CEO and CRO get dedicated help, and the rest of the exec team shares coverage. You can always specialize later — it’s much harder to unwind entitlement once it’s set.  

Steve from Bloomerang:

I used to be very anti-EA. I wanted full control of my calendar, email, travel, all of it. I told myself it was about efficiency, but really it was about trust. Then I met Stacey. She’s my EA now, and she has completely changed my life. I don’t say that lightly. If I ever leave this job, I’m taking her with me.

So I get the instinct to say no when other execs start asking. You’ve gone from zero to one, and now everyone wants their own Stacey.

Here’s how I think about it: Start by budgeting for shared support across execs. At a 400-person company, that often looks like 1 EA per 2 or 3 execs. It forces prioritization and avoids a land grab. Let the CRO’s EA pilot shared support for calendar management, travel, expense reports. Then evaluate who actually needs the help versus who’s just feeling FOMO.

And yes, encourage AI first. Calendar management, note taking, email drafts. If someone can’t figure out how to delegate to AI, they probably aren’t ready to delegate to a human.

Bottom line: if an EA can help your execs spend more time on the hard stuff, it’s a good investment. Just don’t expect everyone to be a Stacey.

PS. Never ask your significant other to coordinate with your EA. I know from experience. 

Wishing you accept a term sheet from someone you actually like,

CJ