Online advertiser DoubleClick (NASDAQ:DCLK) celebrated the revival of the online advertising market with across-the-board improvements over its first-quarter 2003 numbers. Revenues, generally accepted accounting principles (GAAP) earnings, and diluted earnings per share all climbed. Revenues jumped by 13%; that and the diluted earnings rose much more precipitously, due in large part to onetime items (a distribution from a liquidated company and a reversed reserve against losses from an acquisition).

Here's how the numbers broke down: The 13% increase in revenue yielded earnings growth of an astounding 749%, but diluted per-share earnings appear to have risen barely more than half that amount, from $0.01 to $0.05 per share. When you see a result like that, especially at a tech company, your first instinct is probably to accuse: "There they go again, diluting the bejeezus out of out of the long-term shareholders."

And while that is all too often the case (DoubleClick did indeed experience some share dilution), it was not nearly on the scale necessary to explain the divergence between net profit growth and profit per-share growth.

Thus, DoubleClick becomes a good lesson in why it is best to think of those little ticker symbols scrolling by on the Nasdaq board as pieces of companies rather than as, well, little ticker symbols. You see, per-share earnings are great for calculating P/E ratios and provide wonderful fodder for financial writers: "Intel (NASDAQ:INTC) missed earnings by a penny on Tuesday" (or did it?), "Cisco (NASDAQ:CSCO) always beats earnings by a penny," and the like.

But the problem with reducing a company's earnings to its earnings per share is that a lot of rounding goes into the process. In DoubleClick's case, the significant rounding happened in the first quarter of 2003, when the company's $906,000 in profits was divided by its 138,760 weighted average shares outstanding and was reported not as earnings of $0.0065 per share, but as $0.01 per share -- a 35% difference!

Hence the discrepancy in per-share earnings growth. Absent rounding, the company's per-share earnings grew not from one penny to five pennies (for 400% growth), but rather from 0.65 penny to 5.45 pennies (for 738% growth). That's much closer to the 749% growth in the company's earnings.

(And to answer your last nagging question: Yes, the difference between 738% company-earnings growth and 749% per-share earnings growth was indeed due to dilution: 1.7%.)

Tom Gardner can't sleep peacefully at night knowing that there are undervalued, undiscovered companies just waiting to be found for Motley Fool Hidden Gems . To benefit from his restless insomnia, take a free 30-day trial today.

Fool contributor Rich Smith owns no shares in any company mentioned in this article.