What's the best way to compare valuations of Internet stocks? One measure has gained more or less universal acceptance: the ratio of stock price to annualized sales, or revenue per share. The popularity of the PSR (price/sales ratio) reflects investor belief that it's more important for Internet companies to grow revenue than profit, and that revenue is proxy for marketplace acceptance and market share. Without this acceptance and market share, profit potential is limited. PSR creates a common measure for comparing companies and their investors' expectations on that basis.

But PSR is trickier than the more traditional price/earnings ratios. A penny of earnings means essentially the same thing regardless of industry or sector. Revenue, however, demands closer consideration of the nature of the basic business. Comparing Yahoo (YHOO) to Amazon.com (AMZN) and to America Online (AOL) offers an illustration -- and suggests a better measure.

GROSS PROFIT
Yahoo trades at 130 times sales. But few would say that Amazon, which trades at 20 times sales, is only one-third as richly valued as Yahoo.

Retailers' revenues are different from those of service companies. Amazon immediately pays out of revenue the cost of merchandise and shipping. Yahoo doesn't itself sell merchandise. Its revenue comes mainly from selling ad space to those who do. The incremental physical costs of supporting additional content, advertisers, and usage on a Web site is small compared to paying for books and CDs. It's similar to the software business, where gross margins, such as those of Microsoft (MSFT), are 80 to 90 percent. The cost of stamping and boxing CDs is a small part of product price.

That perspective makes it clear why Yahoo's PSR is six times that of Amazon. Though Amazon's $253 million sales in the last quarter were three times Yahoo's $76 million, gross profit was much closer -- $53 million vs. $69 million. From that point on, the types of activities and expenses of Amazon and Yahoo are nearly identical: marketing, product development, and administration.

Much of Yahoo's higher PSR essentially rebalances the lopsided revenue ratio by compensating for the fundamental difference between Amazon's and Yahoo's businesses.

And in practice, price/sales ratio really is a measure of gross margin leverage. So a more direct starting point for comparing e-tailers to Yahoo et al. is a price to gross profit ratio (PGPR).

Using price/gross profit instead of price/sales narrows the valuation gap of Yahoo vs. Amazon to 30 percent from 85 percent. Given the volatility of Internet stocks, that 30 percent difference can be considered nearly a dead heat.

Comparing valuation measures:

AOL VS. YAHOO
It is often observed that AOL's price/sales ratio is only one-quarter that of Yahoo (currently 30 versus 130). That seems to say AOL stock is the better value. But look more closely.

Commerce-related advertising and direct marketing are the chief sources of profit for both Yahoo and AOL. Yet Yahoo's gross margin is 90 percent and AOL's is only 38 percent. That's because advertising is 100 percent of Yahoo's revenue but only 19 percent of AOL's. Eighty percent of AOL's revenue comes from providing dialup access, which has much lower gross margins due to the costs of telephone lines, modems, and customer service personnel.

As with the Yahoo-Amazon example, price/gross profit filters out major irreconcilable characteristics of the basic businesses, enabling comparison that is closer to apples-to-apples than provided by price/sales. On that basis, Yahoo and AOL are dead even.

APPLES + ORANGES
Finally, here's another method of evaluating AOL on comparative price/sales basis. (For now, forget about price/gross profit).

Normally, price/sales for AOL is based upon AOL's total sales, or revenue. But as we noted, AOL's revenue comes from two very different businesses -- dialup access and advertising/commerce. This makes it difficult to use PSR to compare AOL to pure plays in either business.

However, AOL's financial reports break out revenue according to access and ad/commerce. This enables an apples-to-apples comparison -- but with a twist.

MindSpring (MSPG), the most richly valued pure-play consumer ISP, trades at 16 times revenue. So we can apply MindSpring's PSR to AOL's dialup access revenue, and Yahoo's PSR to AOL's ad/commerce revenue. Then we add the two results and divide by AOL's number of shares outstanding.

Combining valuation metrics for AOL:

This shows that if each of AOL's two main revenue streams are valued at the same multiples as the most highly regarded pure-play counterparts, MindSpring and Yahoo, AOL is worth $100 per share.

Of course, the exercise shouldn't be taken per se as a measure of value, but as an element of perspective.

THE PSYCHE OF STOCKS
In addition, simply comparing Internet stocks against each other says nothing about intrinsic value. When there's a major correction in the overall stock market, or just in the Internet sector, net stocks that appear undervalued relative to their peers can be hammered just as much, if not more.

Most importantly, the news and psychology that so heavily influences Internet stocks can swamp empirical measures of valuation. So one Internet stock appearing undervalued or overvalued relative to others doesn't mean as much as with traditional stocks. And, of course, as with all stocks, appearances are just a starting point for analysis and selection.