I knew it.

The business was flawed. The numbers were trending in the wrong direction. The lights were flashing red. This stock was going to get hit.

But I didn't pull the trigger. And when AsiaInfo-Linkage (Nasdaq: ASIA) reported a terrible quarter last week and crashed, I missed out on a 26% one-day gain!

A dog with a very big flea
Let's start with the business. AsiaInfo provides software and IT security products to telecommunications companies in China. Thanks to the rapid expansion of China's mobile phone market, AsiaInfo's sales have grown at a 30% clip over the past three years, while profits have more than tripled. Sounds promising. That is, until you realize that China Mobile (NYSE: CHL) accounts for 70% of AsiaInfo's revenue.

Why might this be a problem? After all, China Mobile is the clear leader in the space, with the largest market share, service network, and brand following.

China Mobile uses the sub-standard TD-SCDMA technology to power its 3G wireless network, instead of the new WCDMA technology. That puts it at a serious disadvantage in the mobile Internet space, which is expected to more than double in China over the next five years. In a recent survey, 58% of Chinese customers picked WCDMA phones as having the best 3G network in the country, versus just 4.3% who picked TD-SCDMA models.

Even more worrisome, Apple (Nasdaq: AAPL) recently partnered with China Unicom (NYSE: CHU) -- in part because of its WCDMA capability -- to bring the iPhone to China. The global popularity of the iPhone has made it a leader in the mobile handset market. That positions the smaller China Unicom to capture considerable market share from China Mobile.

How did China Mobile respond? It chose to go it alone and develop its own technology -- the curiously-named OPhone. That will be a big investment, and a bit of a gamble. You’ve got to wonder why China Mobile didn’t decide instead to dial up partnerships with Google (Nasdaq: GOOG) or Research In Motion (Nasdaq: RIMM). Going with an Android or Blackberry device would probably require less marketing spend and R&D time, and China Mobile wouldn’t have to build a brand from the ground up.  

All of this suggested to me that, because of its dependence on China Mobile, AsiaInfo's heady growth -- and lofty valuation -- was in jeopardy. And that was before I spotted several troubling red flags in AsiaInfo's financials.

Red flag No. 1: Surging unbilled receivables
Like many companies that work under long-term contracts, AsiaInfo recognizes revenue in advance of billing the customer. Because it's up to management to estimate what percentage of the project has been completed -- i.e. what amount of revenue to book -- this kind of percentage-of-completion approach can sometimes misrepresent the actual growth of the business.

A quick look at the trend in AsiaInfo's unbilled receivables clued me in that the company's impressive growth wasn't entirely up to snuff.

Metric

March 31, 2010

Dec. 31, 2009

June 30, 2009

Dec. 31, 2008

June 30, 2008

Sales

$63.5 million

$76.3 million

$58.6 million

$53.7 million

$42.1 million

Accounts receivable*

$71.2 million

$72.6 million

$64.3 million

$52.0 million

$46.2 million

Unbilled accounts receivable

$76.7 million

$72.2 million

$53.7 million

$28.6 million

$23.7 million

Author's calculations.
*Adjusted for IBM distribution business.

The growth in AsiaInfo's accounts receivable -- which is adjusted to remove receivables associated with a distribution deal with IBM (NYSE: IBM) -- looks steady relative to its sales for the past two years. But look at the growth in AsiaInfo's unbilled accounts receivable. Two years ago, they accounted for about half, but by the first quarter of this year, those unbilled receivables accounted for more than 100% of AsiaInfo's non-IBM related receivables. Therefore, more than ever before, the growth in AsiaInfo's receivables -- and consequently, a huge portion of its sales -- was being driven entirely by management's best guess on the amount of revenue that should be recognized in a given quarter.

Red flag No. 2: Declining deferred revenue
AsiaInfo's sudden dependence on unbilled receivables for its growth was a troublesome sign that management was accelerating, or frontloading, its revenue recognition in order to make earlier quarters appear stronger than they actually were. You could tell that AsiaInfo's business was slowing (and perhaps in need of a pickup) by watching the trend in its deferred revenue line.

Metric

March 31, 2010

Dec. 31, 2009

June 30, 2009

Dec. 31, 2008

June 30, 2008

Sales

$63.5 million

$76.3 million

$58.6 million

$53.7 million

$42.1 million

Deferred revenue

$35.2 million

$45.6 million

$31.3 million

$44.4 million

$35.8 million

Deferred revenue is one measure of a company's backlog. In AsiaInfo's case, deferred revenue hasn't kept up with sales, suggesting that the company is either recognizing revenue more aggressively (and depleting it), or that AsiaInfo's business is facing a potential slowdown in future quarters.

Red Flag No. 3: Unwarranted spikes in profit margins
While percentage-of-completion accounting can lead to accelerated revenue recognition by increasing revenue booked in advance of billing the customer, it can also overstate profit margins. Like unbilled receivables, management has wide discretion when estimating the profitability of projects that are still in the process of being completed.

Not surprising, as AsiaInfo's unbilled accounts receivables surged, so did its profit margins.

Metric

March 31, 2010

Dec. 31, 2009

June 30, 2009

Dec. 31, 2008

June 30, 2008

Gross margin

61.5%

57.7%

49.1%

53.5%

45.3%

Operating margin

22.9%

21.5%

10.9%

14.5%

8.5%

Is it really plausible to think that AsiaInfo's operating margin could double from a year ago, and nearly triple from two years ago? Sure, maybe business was a bit better, and maybe AsiaInfo was a more diligent on the cost front. But it's also highly possible that AsiaInfo used those unbilled receivables to juice its top line without a corresponding (and appropriate) increase in expenses, providing an artificial and unwarranted boost to its profits.

Foolish bottom line
If you ask me, I think AsiaInfo was playing an all-too familiar (and risky) game. It was booking higher revenue and fewer expenses in earlier quarters at the risk of facing exactly the opposite in the future. And its business was already under the long-term threat of a slowdown from its most important customer, China Mobile. In most cases, this all leads to one thing: a major earnings disappointment like the one we had last week.

I spotted all the warning signs yet failed to act. By not shorting AsiaInfo, I missed out on a big gain. But I won't be missing many of these great shorting opportunities in the future, and neither should you. You certainly don’t want to be holding the bag when the next AsiaInfo blows up.

Accelerating revenue recognition is a danger sign for investors and one of the five red flags I highlight in a new report I’m working on right now. Click here to get it sent right to your inbox. If you share my concern about the low quality of reported earnings -- or if you are looking to short individual stocks for big gains -- enter your email in the box below. I’ll also send you my latest earnings quality research the instant it is published -- also FREE.