Dealmaker’s Diary: This $500M Company Profits From EVERY AI Device Made

Alpesh Patel|June 19, 2025

Every smartphone. Every robot. Every smart factory.

They all need one thing to work: power modules.

And there’s a $500 million company that supplies them to everyone.

Think about it. Every AI device that gets plugged in needs specialized power management. From edge computing in your car to industrial robotics in manufacturing plants.

Our Stock of the Week has been quietly building the infrastructure that makes it all possible. While tech companies burn through billions developing software, these guys collect steady revenue every time someone builds an AI device.

The proof is in the numbers:

• 57% gross margins (real pricing power)
• Top-quartile cash returns on invested capital
• 32-year track record of consistent performance

Sometimes the best businesses are the ones doing the essential work behind the scenes.

Click on the image below for the name and ticker.

Transcript

Hi, friends. Welcome to the Dealmaker’s Diary. I’ve got one that has been on my radar for a while. I could have actually given this as the Stock of the Week much sooner, because it’s one I’ve liked for a while now.

And it is, in case you don’t know, Monolithic Power Solutions (MPWR).

It’s a company that many of you might not have heard of. In some regards, it’s small because quarterly revenue is just half a billion, which compared to some of the companies we look at is small.

However, they’re in industrial automation, automotive, and consumer electronics. Why would this do well?

This is the thing.

For me, it’s not about trying to necessarily understand what they do. Cloud computing sounds interesting. Consumer electronics, not so much maybe. No. We’re not looking at narratives. We’re looking at numbers. Before we look at the numbers, I want to show you something. Oh, by the way, gross margin is 57 percent – they’ve got pricing power. They’re not about to be affected by the uncertainty that’s going on in the world.

What I wanted to first show you before we look at the numbers, as you know, what I like to do as part of education and information, is just show you how companies are using AI. And in the case of this one, they’re supplying AI as part of their hardware in their power modules.

They’re working with companies who are at the forefront of AI, so they’re benefiting from secondary benefits of AI. If you see what I mean, let me put it this way: I use Microsoft ChatGPT, or Copilot.

So my business benefits. Now I’m not inventing AI or creating or innovating, but I’m using it, and that gives me a secondary benefit of productivity to my business and then some, I should say. High exposure to edge AI growth, explaining how they’re using it in robotics and Internet of Things as well. Computing power should help them, and internally, the company’s doing it.

Now more than normal for a company, they’re relying on other partners and their strength in AI to benefit them. And I don’t have a problem with that. That’s like me saying I’ve hired McKinsey to boost my company. I’m not McKinsey, but they’re helping boost my company.

So it’s similar to that. Good. Fine. Box ticked.

Look at these numbers. There are a few numbers that really are positive for me. First of all, on my value growth income rating, seven out of 10.

Anything above a seven ticks my box. Seven or higher ticks my box. In other words, the valuation of the company, the growth of the company, dividend yields – we look at all of those factors across 10,000 companies, and we say, look, minimum, it’s going to meet this criteria just to avoid picking any companies that might be overvalued or not growing.

Before I come to this, cash return on capital invested is one of the highest. If you’re in the top quartile, Goldman Sachs Wealth Management suggests that as a basket of stocks in the top quartile, you get 30 percent per annum. Not every year, not every stock in the basket, but box ticked. I would have liked the Sortino ratio, which is average return to risk, to be a bit higher, but I don’t mind given that number.

Now it is a bit pricey. You’re paying $40 for every future dollar of profit.

And that is a bit pricey even for a tech company. If we were to even put it in that category. So there’s no getting away from that. Volatility is not too bad. I can live with that. Below 20 percent, I’m OK with.

You’ve got this general direction of travel on the chart. If we get this, which seems plausible and reasonable, that’s the returns you’re looking at. We get this going back up there. That’s what you’re looking at.

Now I wish life was that simple. I could draw a line and move heaven and earth. More likely is you’re going to get this. You are going to get a bit of a move up and then you’re going to be underwater.

Guaranteed, this is not going to be the low of the next 12 months. Just because it moves, we know it tends to move. Here’s a little trick I do. I tell people, look at the historic price moves, and then look at the drawdown depth – peak to trough – and the duration until it recovers.

The depth and duration tells you your risk. Doesn’t mean it’s going to repeat itself, but tells you your worst-case risk. And then you decide, well, what kind of guy am I? You might say, well, I’m the kind of guy whose duration doesn’t ever want to be 12 months.

I’m the kind of guy whose depth doesn’t want to be more than 25 percent. In which case, now you know. You take that little template and you say, well, I don’t want this stock. I don’t want this.

I don’t want this. It also helps manage your expectations and keeps you calm because the two reasons people mess up in the market are panic – drawdown – and boredom – duration.

So we are with that one. Like I said, it’s overvalued on a discounted cash flow basis as well, and that’s the big problem there.

But aside from that, direction of travel is good. So I hope you liked it. Thank you very much.