I am a bona fide telecom junkie. Nothing fascinates me more than trying to find the best telecom investment, especially since prices have been beaten down since 2000. Despite its gyrations, it's a vital industry to our country and to the world, and my hunch tells me some company has to come out a winner eventually. And I have always wanted to figure out which one that was going to be.

Level 3 (NASDAQ:LVLT) was going to revolutionize the wholesale data transmission industry by building the best fiber optic network. With so much pent-up demand that could materialize because of projected price elasticity, Level 3 was supposed to be one of the most influential and successful businesses in telecom. But I just couldn't pull the trigger. While the story was compelling, I kept reminding myself that it was just a story.

Qwest (NYSE:Q) was going to be a be-everything-to-everybody telecom company. It would offer local, long-distance, data, and wireless service to consumers and businesses, and via wholesale channels as well. Qwest would have a diverse group of revenue streams to help cover the fixed costs of its network. And in July 2002 I bought in at $1.62 per share. The reason -- the only reason -- I bought the shares was that I thought "$1.62 just seems way too low for this company."

Six months, six dollars
Six months later, the stock was up to $6 per share. Clearly, my hunch was correct, I thought. It was too low, and the market responded. I kept thinking that I should probably do some deeper analysis. But with my return approaching 300%, I got complacent and just let it ride. And ride it did -- right down from four-bagger status to below two-bagger territory. Easy come, easy go.

I made a mistake
 Last month, I revisited my Qwest purchase. Although I hadn't spent a ton of money originally (it was only some money left over from nonreinvested dividends), I couldn't help but feel foolish about why I'd bought the stock -- I'd been so enamored by Qwest's potential that I hadn't taken an emotionally honest inventory of all the criteria afloat in its seas. And ultimately, I'd bought on a hunch -- one that didn't have much analytical buoyancy at that.

Sunlight is the best disinfectant
And speaking of aphorisms, we all know a mistake is only one if we fail to learn from it. My little hunch captured some good things -- that sent the stock up -- along with some bad things -- that sent it back down again.

The good
Like any responsible CEO, Qwest's Richard Notebaert has been working on his plan for his company's future.

He's wanted to reduce interest expense, and he's done so by restructuring and retiring debt. He sold Qwest's directory business, QwestDex, in August 2002 to help get the cash situation under control. Since then, Qwest has also sold its wireless assets to Verizon (NYSE:VZ).

Notebaert also recognized that voice over Internet protocol (VoIP) is the wave of the future and has aggressively rolled out the service.

The bad and the ugly
Qwest still can't raise wholesale prices associated with competitors using its network. That means a competitor such as MCI (NASDAQ:MCIP), which recently emerged from bankruptcy and is a Motley Fool Inside Value pick, continues to have low-cost access to local area networks that it uses to sell long-distance service.

Second, Qwest is losing customers. An August 25, 2004, article in The Wall Street Journal mentioned that Qwest has lost about 3 million lines since 2000, including 200,000 last quarter.

Third, there is intense price competition. Everyone wants traffic to go through their network because the networks have huge fixed costs. In addition, networks are non-redeployable assets -- meaning they can only be used as a network. At least with a factory you can retool if you must. So, rather than close up shop, companies drop prices, which in turn makes generating profits more difficult.

Fourth, everyone with a network or access to one is offering VoIP. Cable companies such as Comcast (NASDAQ:CMCSA) now offer phone services across their network and bundle it with cable and broadband. Vonage just uses the public Internet to send voice. Barriers to entry are practically gone now that the disruptive technology of VoIP is a reality.

Fifth, Qwest still has huge network maintenance expenditures to deal with month after month. And if you add the fixed cost of debt to the fixed cost of maintenance, that spells trouble if revenues decline. Remember what happened when US Airways took on too much debt, passenger seat miles dropped, and they couldn't raise prices?

Finally, Qwest has legal and settlement fees to deal with because of past issues like improper accounting. In its second-quarter 10-Q (see Note 12), Qwest increased its legal reserve fund to $500 million by adding an additional $300 million (note that the company expects to recoup most of these funds by hitting up its insurance companies).

I'm selling
 The competitive environment stinks. No one can create a competitive advantage because everyone essentially competes the same way. Without any differentiation, the winner will be the lowest cost producer. But even then, the competition will likely erode the business value to the point that the only eventual winner ends up being the customer.

So just as I'm going back on my once-coveted hunch that a real winner will emerge in telecom, I'm correcting the mistake I made when I bought Qwest on a hunch in the first place. I'm selling because I can't make a reasonable valuation. I'm selling because customers are leaving. And revenues are declining. And legal fees are growing.

When you can't make heads or tails of the business environment nor a reasonable valuation, you can't make a stock purchase. And since I wouldn't buy Qwest today, I can't justify holding it either. That's really all there is to it.

Do as I say, not as I did
From all of this, please remember three things:

1. Be critical when you read about a "good story."
2. Know the reasons why you are buying.
3. If you've got the experience, try to make at least a rudimentary estimate of the company's value.

This last point is admittedly the hardest, but I'd encourage you to, over time, learn a little bit about the forces traditionally at play in valuation. You don't have to be an expert to see that, for example, a stock whose current price seems predicated on triple-digit earnings growth for the next 20 years is going to have a tough time filling its shoes.

Just because you buy a stock and it goes up doesn't mean you're doing anything right. I made a guess and got lucky. Over the long run, that's not the Foolish way.

The Motley Fool is committed to bringing you the best investment ideas. With investment newsletters like Income Investor , Hidden Gems , and Inside Value , you get solid analysis, not pie-in-the-sky wishes. So take a free trial of any of them today.

Fool contributor David Meier owns shares of Qwest, even though he is not proud of it. He will be selling his shares under the guidelines of the Fool's disclosure policy. He does not own shares in any of the other companies mentioned.