"Month-end close is a ball and chain. It keeps us all where we are."
That’s a real finance leader describing her close. Every month, it drags the whole team back to the same starting line, rebuilding workpapers they already built.
Ledge gives every close task its own AI accountant. It connects to NetSuite, banks, payroll, and 150 other systems. It pulls the data, builds the workpaper with live formulas, drafts the journal entry, and posts it through your approval workflows. Day one of close, your team opens reviewed work instead of blank tabs.
✍️ If you’ve got a tricky CFO grade problem in your day job, and you want my help… you can send me your questions anonymously, and you might get features in next week’s Mailbag. Just submit using the button below:
Lots of fun stuff on tap today, with 3 interesting questions from CFOs with a problem to solve... here’s what’s up:
Why you shouldn’t make the case for AI ROI yet
Treading the fine line between the CEO and investors
The right way to let AI into your inbox
Now, let’s get into it.
Being_CFO_is_a_blast from Switzerland asked:
One of the least rewarding tasks I have found myself doing in the past, was building an performative business case in Excel to justify an investment in a resource. You know, the type of circular business case that simply assumes a notional revenue boost from hiring one extra marketing person.
With the current wave of "show me the AI productivity benefit" requests, what advice do you have? What is the thinking CFO's solution to business case circularity? How can you apply it in reverse to the cost avoided by AI questions?
We are cut from the same cloth here. I despise performative business cases for things that do not merit them. I call it ‘false financialization’. It ties up valuable internal resources and focus, all to manufacture numbers, simply because an exec wants the cover of a "data-based decision". Even when the data is incomplete, which on these kinds of decisions it almost always is.
Reminds me of one of my favorite quotes:
If the data and the anecdotes disagree, the anecdotes are usually right.
That is a far more elegant way of making the same point. One of many reasons he is a billionaire and I am not. Although I do have hair…
AI has this problem on steroids. People scrambling to justify investment with elaborate business cases, when the truth is they have no idea. The models and the capabilities are moving so fast, in both directions, that any precise forecast is essentially fiction. My view: In the short run, people will overestimate the benefits. In the long run, they will dramatically underestimate them.
So here is how I would actually approach it.
Focus more on measuring what happened, rather than trying to predict what will. With AI specifically, what happened when you tried something is far more valuable than what you think might happen. No-one has a f*cking clue how to calculate payback on AI spend yet, and the most dangerous people are those who are adamant they do…
If you can’t write a decent business case, you are by definition in ‘discovery’ phase. So your conviction and investment in AI has to match that uncertainty (i.e. small and experimental). And you can scale investment and focus as your understanding, data and success criteria grows…
But it’s a great question, AI spend is genuinely hard to measure and control, so think about that discipline upfront. I wrote a full piece on it here.
TLDR: Don’t forecast AI benefits you cannot possibly know. Fund wide, cheap experimentation to find the real opportunities, then take one or two big, well-resourced swings.
Something Better from Ireland asked:
Dear Secret CFO, thanks so much for your sage advice delivered weekly. I'm a first-time CFO of a PE-backed software company. I was hired directly by the PE firm into a brand new CFO role just over four years ago. The founder CEO exited two years ago and was replaced by a new CEO.
We went through a DD process in the last 12 months. Early on, only the CEO, myself, the advisory firm, and the PE owners were involved, with the wider SLT brought in later. It fell through.
Over the last few weeks, our CEO has mentioned on occasion that they are in contact with the founder of a company they previously worked at, with the view of that company acquiring our business. Our CRO, who joined after the CEO, also worked there. They operate in our industry but aren't a competitor. The CEO says they, our investors, and the same advisory firm have all had calls with this company. Nothing is signed yet, pending the acquirer's funding.
Here's where it gets off. The CEO told me our investors asked them not to discuss any of this with me, and that I'm the only person who knows. Yet I know the CEO is sharing our financials with the acquirer, because I recently prepared DD-format financial data for exactly that purpose. And expenses on the CEO's card show that they and the CRO had an in-person meeting with the two founders of the interested company, plus a senior exec from a business that company recently acquired.
So, I'm being told this is top secret and that I'm the only one who knows, but the CRO is clearly in the room too. I find the whole thing a bit off and don't know how best to approach it. Would appreciate your thoughts.
Hmm. This is a tricky one.
CFO roles, and PE-backed ones especially, are full of moments where you are caught between your CEO and your investors, unsure who knows what. They require careful treading. The precise answer is always situation-specific, and I will come to yours in a minute, but first some general principles that hold up across most of these situations.
Be a force for transparency: If both parties trust you, use that to gently encourage more open communication. It rarely helps in the short term, but over time you can make a real difference by giving each side confidence that the other can be trusted with the information they treat as sacred.
Never go behind your CEO's back unless absolutely necessary: It is a trust killer. There are times when integrity demands it, but treat it as a one-way door. A nuclear option, not a tactic.
Know your boundaries: As you gain experience at this level, you will get clearer on what you are and are not comfortable with. You have fiduciary responsibilities, and you need to understand where they sit relative to your loyalty to the CEO.
Keep plausible deniability: If you are asked not to discuss something, and you are comfortable with that, make sure you stay in a position where you can honestly say "I don't know."
Never lie: If the party out of the loop asks you outright, do not lie to protect someone. It sounds obvious, but these questions come out of the blue and you have no time to think, which is exactly how accidental lies happen. If your investor says: "I heard Janet spoke to so-and-so, is that true?,” you do not confirm or deny. You simply say: "I think you should talk to Janet about that."
Now, to your situation. You essentially have two scenarios.
One: your investor is in the loop and, for some reason, does not want you to know. Possible, but it seems unlikely given what you have described.
Two: your CEO is making kissy faces at their old company behind your investors' backs, and is worried you will tell them. Based on the facts you have shared, this looks far more likely.
Honestly, it still sounds early. You have shared some previously prepared information, presumably under a NDA, with a contact your CEO and CRO know from a former employer. It might be something. It might be nothing. It might just be your CEO enjoying the chance to swing their private parts around and show an old contact how important they have become. I have seen that more than once.
But I would also say: most M&A deals I have been involved in started with exactly this kind of informal contact. It could also be the first stage of a serious strategic exit.
Here is what I would do.
Go to your CEO and say something like this: "I know you have had a few meetings about possible M&A. I have not asked how they went, because I do not know who is in the loop and who isn't, and I want to keep plausible deniability if I am asked. Particularly with our investors; but you said they were across this, is that right?"
That does two important things.
First, it makes clear that you currently feel out of the loop, and that is exactly what you will say if anyone asks you.
Second, it forces the CEO to answer the question about the investors. Either they confirm it, or they lie to you. And if it turns out the investors were not in the loop, you are protected: you were explicitly told they were, and that the CEO was handling communications. You can say that to the investors later with a clear conscience.
Then put down a marker. Tell the CEO that if these conversations develop, and you start being asked to prepare specific deal information, you will want that discussed at board level before you go any further. That sets a clear guardrail, and it is very hard for the CEO to object. After all, they did say the investors were in the loop…
Get comfortable in these gray areas, my friend. This will not be your last time.
TLDR: Probably early-stage and possibly nothing, but protect yourself. Tell the CEO you feel out of the loop, make them confirm the investors know, never lie if asked, and set a clear marker that any real deal work goes to the board first.
Hunter from Canada asked:
You attributed some of your success to your email writing/communication. Nowadays many rely on AI to write their emails, draft their narratives, and as a result, some of the muscle of communication isn't getting exercised. Tools like Wispr Flow make it easier to not filter the inputs.
While AI adoption is great, how would you advise the next generation of CFOs to leverage the tool but not lose their ability to communicate?
Thanks Hunter. You are referring to my Mailbag answer from a couple of weeks back.
And yes, you have put your finger on exactly the right trade-off. This is about more than emails though. It is about knowing when to let AI do the work for you, versus when to use AI to help you do the work yourself.
The analogy I use is this. Imagine you have a 50 pound boulder you need to get to the top of a hill. If I offered you a robot to do that for you, at no cost, would you take it? The answer should be yes. The job is to get the boulder up the hill, and you should want to do that with the minimum effort possible, so you can spend your time and energy elsewhere. Even if that is just shifting more boulders.
Now imagine you are standing in a gym above a 50 pound kettlebell, about to start a set of 12 swings. If I offered you a robot to do that for you, again at no cost, would you take it? The answer should be no. The job here is to build strength, not to move the kettlebell, and a robot cannot do that for you.
Bring that back to your email question, and you can use Daniel Kahneman's fast thinking (System 1) versus slow thinking (System 2) in a similar way. Triage the emails you need to send into one of the two branches.
System 1 is for the high volume of low complexity emails. Scheduling, confirmations, routine replies. The kind where the difference between a good and a bad version is negligible. For these, get your AI tool trained on your voice and let it draft them in batches for you to review and fire off. The whole point is speed. You have already triaged them as low risk, so treat them that way.
System 2 is for the emails that carry weight. Where context matters, and where the gap between a great message and a mediocre one is significant. Your month end performance commentary to the exec. A response to a tricky investor question. A kick in the ass to an underperforming department. Here the rule flips: you hold the pen, and AI helps you. Not the other way round.
I use AI in my own writing now. Researching a specific point, fact-checking, running edits, unlocking a sentence I am stuck on (bearing in mind every sentence I write consumes roughly 40 hours of global finance leadership time, I take this seriously). But the crucial distinction is that AI is never thinking for me. It is helping me think better.
Used well, AI is like a focused isolation exercise, targeted reps that build the muscle. Used badly, it is letting a robot lift the weight while you watch, which builds nothing. The goal is sharper thinking and clearer communication, not outsourced thinking.
Practically, I run a set of custom Claude and ChatGPT bots, all explicitly instructed to draw understanding out of me, provoke me, challenge me, and edit me. Never to generate ideas or text from scratch. Another powerful move is to have AI tear down your argument, or feed it the context of your recipient and ask cold what it would do on receiving your draft.
TLDR: Let AI loose on low-stakes emails at speed. But for the ones that matter, hold the pen yourself and use AI to think better and faster, but never to think for you.
A few of the biggest stories that every CFO is paying close attention to. This is the section you might not want to see your name in.
CFO pay is up 8%. LTI awards up 12%; outpacing CEOs. I guess the next coffee round is on us…
It’s nice to see a CFO transition be so well planned, rather than someone getting parachuted in, in response to a crisis.
Eesh… being named in a criminal case stuck between the politics USA, Iran and China. Honestly… not on my CFO bucket list.
ICYMI, here are some of my favorite finance/business social media posts from this week.
It’s been hard not to enjoy the scenes of the Scotland fans in Boston for the World Cup over the past two weeks:
High yield bond investors have ‘trigger words’. I unknowingly used one on an earnings call a few years back… it’s a story for another day. But this made me smile:
If you’re looking to sponsor CFO Secrets Newsletter, fill out this form, and we’ll be in touch.
If you enjoyed today’s content, don’t forget to subscribe.
You can help make sure this newsletter always stays free simply by spreading the word. And when you share CFO Secrets with your finance friends, you’ll earn rewards, including a 50-page PDF guide on what it takes to be a great CFO. Start sharing your unique referral code today: {{rp_refer_url}}
Let me know what you thought of today’s Mailbag. Just hit reply… I read every message.
Last weekend’s Playbook laid out the unit economics of growth and the importance of contribution margin.
Last week’s Boardroom Brief asks why so many CFOs are burning out now.
Disclaimer: I am not your accountant, tax advisor, lawyer, CFO, director, or friend. Well, maybe I’m your friend, but I am not any of those other things. Everything I publish represents my opinions only, not advice. Running the finances for a company is serious business, and you should take the proper advice you need
