Don't go setting any more $400 price targets on

Amazon.com

(AMZN) - Get Report

just yet.

Shares in the company have jumped 25% since a Thanksgiving weekend

deluge of data that bulls say point to unexpected sales strength. Traffic over the weekend at Amazon.com jumped 33% from a week ago, according to one widely quoted study.

But a fund manager and a research firm with a solid track record on reading the Amazon tea leaves say the trends aren't so cut and dried. In fact, this analysis suggests there's a good chance Amazon won't meet Wall Street's fourth-quarter expectations, let alone trounce them. That's crucial because Amazon has designated this as the quarter in which it will at last turn profitable by certain limited measures. If it fails to do so, it risks the wrath of uncharitable investors.

Tracking

According to data compiled by comScore, a West Coast research firm that tracks activity in Amazon's core U.S. retail business, fourth-quarter domestic sales through Monday were $345 million. That's about $20 million short of the pace Amazon would need to reach analysts' expectations, according to a fund manager who compared comScore's data with trends gleaned from the year-ago period. Amazon, which lately was off 59 cents at $11.06, didn't immediately return a call seeking comment.

Analysts surveyed by Thomson Financial/First Call expect Amazon to generate revenue of $1.01 billion for the fourth quarter, up from $972 million a year earlier. To reach $1 billion in overall sales, Amazon would need to bring in at least $700 million in domestic revenue for the quarter, going by historical trends, which show that Amazon typically gets 70%-75% of revenue from its domestic business.

But Amazon's selling pace through Monday isn't going to be enough, says Eric Von der Porten, a money manager at Leeward Investments who owns put options on Amazon -- essentially a bet the stock price will fall.

In Too Deep?
Fourth-quarter sales estimates

*comScore estimate. **Ragen MacKenzie.
***Thomson Financial/First Call

Based on Von der Porten's analysis, Amazon is on track to book just $664 million in domestic sales. "Amazon would seem to be running well behind both last year's rate and the expectations for this year," Von der Porten says.

Von der Porten arrives at his conclusion by figuring that last year Amazon had booked some 52% of fourth-quarter sales by Nov. 26, based on analyzing actual sales for the quarter and last year's items-sold data. Considering the weakness of the economy and the soft sales reports issuing from many retailers, it seems only fair to assume this holiday season will follow roughly the same pattern.

So unless sales surge next month, Amazon could face a fourth-quarter domestic sales shortfall of at least $36 million, according to this analysis, whose methods were described as solid by another source who follows Amazon closely.

More than Amazon's fourth-quarter numbers are at stake. Amazon, which came in the dying days of the dot-com bubble to be a whipping boy for the profit-shunning ways of many tech companies, has promised investors it would get out of the red in the fourth quarter for the first time ever, on a limited basis excluding certain expenses. If the company fails to make good on its promise, its standing on Wall Street -- already tenuous because some observers say the company has been less than generous with financial data -- could deteriorate further.

Sample This, My Pit Bull

Von der Porten's analysis of the comScore data departs markedly from Wall Street's reading of Amazon's Delight-O-Meter, a running tab of items sold during the holidays. Citing Delight-O-Meter readings, which Amazon readily admits are unscientific, investment houses including Merrill Lynch, Wit Soundview and Prudential Securities have published reports saying Amazon could top estimates.

In fact, many people have been expecting online retailers such as Amazon to benefit from consumers' fears following September's terrorist attacks. The notion that people will shun the malls and order online has helped Amazon stock to more than double from its preatrocities level.

But ComScore, which has sought to fill the vacuum created by Amazon's reluctance to release monthly sales as traditional bricks-and-mortar retailers do, has been spot-on in the past in estimating Amazon sales. The company has a large sample size of roughly 1.5 million people, and uses its technology to track sales as they happen, rather than asking consumers to remember their purchases after the fact, as other firms have done. In the third quarter, Amazon's $454.5 million in domestic sales nearly matched comScore's estimate of $455 million.

Halftime Show

Of course, nothing is set in stone at this hour. ComScore spokesman Dan Hess cautions that it's early to determine conclusively where Seattle-based Amazon's sales will end up. "It's a bit like trying to call the final score at halftime," he writes in an email.

And even the brokerages caution investors that gauging midquarter sales trends is far from an exact science. Indeed, the firms suggest Amazon shares may be pricey after their recent run; none of the three has a buy recommendation on the stock.

In any case, Amazon has a busy December ahead if it hopes to keep its Christmas promises.