ASML is headquartered 45 minutes from my university in Leiden. Their building is visible from the train. It felt like the right company to practice my first DCF valuation on — a company I can literally see from public transport, that happens to be one of the most important businesses in the world.
This is not investment advice. I'm a data science student, not an analyst. But walking through a DCF from scratch — with real numbers, real assumptions, and real uncertainty — taught me more about finance than any textbook chapter. Here's the process, including every place I was unsure and every assumption I had to make.
Step 1: Get the numbers
ASML's 2024 annual report gives us the raw inputs. I pulled these from their consolidated financial statements:
- 2024 Revenue: ~€28.3 billion
- 2024 Free Cash Flow: ~€9.2 billion
- Shares outstanding: ~393 million
- Current share price: ~€680 (at time of writing)
The FCF number is the important one. That's the cash left over after all expenses, taxes, and capital expenditures. It's the money that theoretically belongs to shareholders.
Step 2: Project future cash flows
This is where the art starts. ASML has guided for €30-40B in revenue by 2025, and the EUV/High-NA cycle suggests strong growth through 2028. But semiconductor cycles are brutal — what goes up comes down.
I modeled three scenarios, because pretending you know the future with one number is dishonest:
| Year | Bear ($B) | Base ($B) | Bull ($B) |
|---|---|---|---|
| 2025 | 8.5 | 10.0 | 11.5 |
| 2026 | 8.0 | 11.5 | 14.0 |
| 2027 | 7.5 | 12.5 | 16.0 |
| 2028 | 8.0 | 13.0 | 17.0 |
| 2029 | 8.5 | 13.5 | 17.5 |
The bear case assumes the AI capex cycle cools and China restrictions bite harder. The bull case assumes High-NA EUV ramps faster than expected and ASML's backlog converts at higher margins.
Step 3: Pick a discount rate
The discount rate (WACC) is supposed to represent the return an investor demands for the risk of owning this company. For a mega-cap with a monopoly-like position, something in the 9-11% range feels reasonable. I used 10%.
For terminal growth (how fast the company grows forever after year 5), I used 3%. ASML isn't going to grow at 15% forever, but it's also not going to stagnate — semiconductors are a long-term growth industry.
Step 4: Run the numbers
I plugged the base case into my DCF calculator (yes, the one on this site). Here's what came out:
| Year | FCF (€B) | Discount Factor | PV (€B) |
|---|---|---|---|
| 2025 | 10.0 | 0.9091 | 9.1 |
| 2026 | 11.5 | 0.8264 | 9.5 |
| 2027 | 12.5 | 0.7513 | 9.4 |
| 2028 | 13.0 | 0.6830 | 8.9 |
| 2029 | 13.5 | 0.6209 | 8.4 |
| Terminal | 13.9 | 0.6209 | 123.4 |
| Total | 168.7 |
Enterprise value: ~€169 billion. Per share: ~€429.
Current price: ~€680. So according to my base case DCF, ASML is overvalued by about 37%.
Step 5: Realise why DCF is both useful and dangerous
Here's what this exercise actually taught me:
The terminal value (€123.4B) represents 73% of the total valuation. That single number — which depends on a growth rate I essentially guessed — determines most of the outcome.
If I change the terminal growth from 3% to 4%, the per-share value jumps from €429 to €528. If I drop WACC from 10% to 9%, it jumps to €563. Small changes in assumptions produce massive changes in output.
This is why Damodaran calls DCF "a tool for structured thinking, not a calculator for truth." The value isn't in the final number — it's in forcing yourself to make explicit assumptions about growth, risk, and time. Once those assumptions are visible, you can argue about them productively.
What I'd tell someone doing their first DCF
- Do it on a company you can physically see. It makes the numbers feel less abstract. ASML's factory is right there — I've walked past it.
- Run three scenarios, not one. Anyone who gives you a single DCF output is either lying or lazy. The range matters more than the point estimate.
- Pay attention to where the terminal value dominates. If 70%+ of your valuation comes from the terminal value, your model is basically a guess about the long-term growth rate dressed up in a spreadsheet.
- Compare your result to the market price. If your DCF says a stock is 50% undervalued, the most likely explanation isn't that you've found a bargain — it's that your assumptions are wrong.
You can run your own version of this analysis at tuixue.nl/fintools. The DCF calculator there is exactly what I used for this post. Enter your own assumptions and see how the numbers change.