Recently, I was reviewing investment portfolios and came across something that many people overlook: most serious investors talk about nothing but the P/E ratio when they choose companies. So I decided to dig deeper into this.

The P/E ratio, or price-to-earnings ratio, basically tells you whether you’re paying a fair price or throwing money away. It’s calculated by dividing a company’s market capitalization by its net earnings. Or, if you prefer it more simply: the stock price divided by earnings per share. That’s the P/E.

Now, this is where things get interesting. A low P/E (between 0 and 10) may look attractive, but watch out: sometimes it means earnings are about to fall. A P/E between 10 and 17 is what most analysts consider the comfort zone, where you expect growth without overpaying. Above 17 up to 25, you start seeing signs that either the company has grown a lot or you’re close to a bubble. And if you see a P/E above 25, well, that can mean the market has a lot of hopes—or that it’s simply in the stratosphere without any justification.

What’s fascinating is that the P/E doesn’t work the same way across all sectors. Look at Arcelor Mittal, which is in metallurgy, with a P/E of 2.58. Then you look at Zoom Video with a P/E of 202.49. Are both overvalued? Not necessarily. Banks and the industry typically have low P/Es, while tech and biotech live in a different universe.

There’s also the Shiller P/E, a more sophisticated variant. Instead of looking only at one year’s earnings, it takes the last 10 years adjusted for inflation. The theory is that with this broader perspective, you can better predict the next 20 years. It makes sense when you consider the volatility of year-to-year results.

Then there’s the normalized P/E, which is even more refined. Here you subtract liquid assets from market capitalization, add debt, and instead of net earnings you use free cash flow. It’s like cleaning up all the accounting grime to really see what’s going on. The case of Banco Santander buying Banco Popular for 1 euro is perfect to understand this: it wasn’t really 1 euro—it was taking on a massive amount of debt that changed the entire analysis.

So what is the P/E ratio actually useful for? It’s a comparison tool. If two companies in the same sector have different P/Es, the one with the lower P/E might be undervalued. But here’s the big caveat: you can’t rely only on the P/E ratio. There are companies with low P/Es that are literally on the verge of going bankrupt. The story is full of cases like that.

That’s why value investors—the ones looking for good companies at a good price—become obsessed with the P/E. Funds like Horos Value International have a P/E of 7.24 compared with 14.55 for their category. That’s not a coincidence.

The weakness of the P/E is that it looks only at one year of earnings, which is pretty short-sighted. It doesn’t work for companies without profits. And for cyclical companies, it’s a disaster: at the peak of the cycle you see a low P/E; at the trough you see a very high P/E, when in reality it’s the same company at different moments.

What I’ve learned is that the P/E is a useful tool, but it has to be accompanied by other metrics: BPA, ROE, ROA, P/VC. Good fundamental analysis requires you to look beyond a single ratio. Look at the business composition, understand where the profits come from, and verify that they’re not just from the sale of a one-off asset.

So when someone tells you a company is a good investment just because it has a low P/E, know that they’re oversimplifying too much. The P/E is the starting point, not the destination. Use it to filter, but then do the real work: spend time understanding the company, its sector, and its competitive position. That’s what separates serious investors from those who lose money.