I remember working on a DCF for a fast-growing medical device company back in my EY days.
Revenue was projected to grow aggressively.
Margins were improving.
The story looked compelling.
But one line stopped me.
CapEx = Depreciation.
It didn’t make sense.
If the business is expanding — how can CapEx stay flat?
That was the moment I started questioning something more fundamental:
Whether our models actually reflect how capital decisions are made in reality.
Why CapEx matters
At its core:
FCF = EBITDA – CapEx – ΔNWC – Tax
Get CapEx wrong, and everything breaks.
Free cash flow gets overstated
IRRs look attractive — but misleading
Investment decisions become distorted
So let me ask:
When you build a model — do you actually know what your CapEx number represents?
Or is it just a simplifying assumption?
Internal reality vs external modeling
Inside a company, CapEx is not a formula.
Decisions are driven by:
ROI / NPV / payback
Capital constraints
Governance processes
Strategic priorities
Companies don’t think in “maintenance vs growth.” They think in decisions under constraints.
Outside the company, we approximate.
Common approaches include:
Linking CapEx to depreciation
Using a % of revenue
Benchmarking against peers
Inferring from asset turnover or growth assumptions
These are useful.
But they remain approximations.
So think about this:
How confident are you that your CapEx assumption reflects what the company will actually do?
This is where most valuation models break.
But the real issue isn’t the formula —
it’s how capital decisions are actually made inside the company.
👉 The gap becomes obvious once you look at real-world CapEx decisions.
What models miss: three realities of CapEx
1. Advisors don’t control CapEx assumptions
In M&A, we often rely on management’s CapEx plan.
In theory, we validate it.
In practice:
DD asks for a “5-year CapEx plan”
Numbers are provided
But the underlying logic is often unclear
Sometimes even internal teams can’t fully explain it.
As an advisor, you can challenge —
but you cannot arbitrarily replace it.
So the model often inherits that uncertainty.
2. CapEx decisions are shaped by factors beyond finance
Large investments go through approval processes.
But DCF is rarely the sole deciding factor.
I remember, during my time in a chemical company, we faced a situation where:
The market was structurally declining
Competition from China was intensifying
From a financial perspective, the outlook was clearly unattractive
Yet significant CapEx continued.
Why?
Because:
Long-standing relationships with key customers needed to be maintained
Exiting would have meant losing strategic positioning
The investment was seen as a necessary cost to stay in the game
In that context, reallocating capital to higher-return opportunities was not considered realistic.
From a model perspective, it looked inefficient. From a business perspective, it was unavoidable.
3. Not all “investment” is CapEx
When I was in a startup, the most important investment wasn’t CapEx.
It was marketing.
Customer acquisition.
Growth.
Market capture.
Internally, we often discussed:
CapEx
Marketing spend
Growth investment
…as part of the same capital allocation problem.
Yet in financial statements:
One is CapEx
One is Opex
So here’s the question:
If two types of spending drive growth in similar ways — should they be treated differently in valuation?
Accounting says yes.
Economics is less clear.
Closing
Most models look clean.
Reality is not.
So the real question becomes:
Are you modeling reported numbers — or trying to understand actual capital decisions?
One last question
How do you deal with CapEx assumptions in your own models or decisions?
There’s no perfect answer.
But the difference between assumptions and reality is often where valuation gaps come from.
If you’re working on valuation, DD, or capital planning,
this is where most models quietly fail.
Dai Kadomae, CFA, CPA GARYO FINANCE | LinkedIn