Before we get to the meat of this column, a few housekeeping tasks. Northrop Grumman (NYSE: NOC) reported its fourth-quarter and full-year 2007 earnings last week. Profits grew 16% year over year in 2007, and 2% in Q4 versus last year's comparable quarter. Sales rose 6% for the year, 10% in Q4 alone. Kudos.

And if you want to read more about the news, per se, I'm sure you can find plenty of fascinating regurgitations of the press release.

Thinking deeper
Me, I can write about earnings news any time I like. But I think it's more interesting -- and potentially more lucrative -- for us to take a step back today from the incessant march of earnings season and instead give some thought to what the numbers mean.

As you may recall, when discussing the earnings news at defense contractors General Dynamics (NYSE: GD) and Lockheed Martin (NYSE: LMT) last week, I talked a lot about backlog. I expect that, if the news warrants, I'll talk more about backlog when looking at Raytheon (NYSE: RTN) and L-3 (NYSE: LLL) this week. But I haven't yet really laid out the reason why I'm talking so much about this subject. Today, I'll be using Northrop's earnings release to dig deeper into the subject of backlog, and how an investor can use this single line of data from an earnings press release to help divine where a company might be headed in the future. Let's start with the theory.

What is backlog?
Backlog refers to orders that have been booked and are expected to become revenue but haven't yet. Backlog is also a dynamic number: As existing orders are filled, they exit backlog and become revenue. And as new orders are booked, they enter backlog. (Except when they don't. As I recently learned in an interview -- read the whole thing or just a taste -- with American Science & Engineering (Nasdaq: ASEI), some orders get filled so rapidly that they never show up in backlog. They're received and filled within the space of a single quarter).

Backlog can even "evaporate" when a customer changes its mind about an order it's placed -- a situation we have seen with Airbus and its much-delayed A380 flying barn and could soon see at Boeing (NYSE: BA), if it doesn't get its act together with the 787 Nightmare Liner.

Northrop as guinea pig
Let's "back out" the variables of quick-fill orders and evaporating backlog, though, and see how the basic order-backlog-revenue situation should work, using Northrop as an example:

  • At the end of fiscal 2006, Northrop had $61 billion in "total backlog" (funded and unfunded contracts.)
  • At the end of fiscal 2007, we now know that Northrop's total backlog increased to $64.1 billion.
  • We also know that over the course of fiscal 2007, Northrop booked $32 billion in revenue.

Logically, therefore, what seems to have happened over the course of fiscal 2007 was this: Northrop started with $61 billion in orders waiting to be filled. It filled more than half of those -- $32 billion worth, to reduce its backlog to $29 billion. But it also received new orders worth $35.1 billion, to leave it with $64.1 billion in backlog at year's end.

Now, if we back up a bit, we can see what could have happened to Northrop last year.

  • Its backlog could have gone down. That would have meant that Northrop sold so much stuff over the course of the year that it ate into its existing backlog of orders to sell stuff in the future. If backlog represents future sales, that would have suggested that in the future, sales were bound to decline.
  • Its backlog could have remained unchanged. That would have meant that Northrop sold precisely as much stuff, as it took in orders for new stuff. We could thus infer that Northrop's future sales would stagnate -- not rise or fall, but remain steady.

What did happen
And then there's what did happen. Backlog increased by $3.1 billion, or 5%. To me, that portends continued sales growth for Northrop. However, it's worth underlining that sales grew 6% in 2007, so backlog, or future sales, is growing less quickly than actual sales grew last year. Logical conclusion: Northrop will continue growing in 2008, but perhaps not as quickly as it did in 2007.

Of course, that's just my opinion
Let me close this column with a disclaimer I borrowed from HBO funnyman Dennis Miller: "That's just my opinion. I could be wrong."

The idea that we can use past trends in backlog to predict future trends in sales is only a theory -- it won't become fact until we put it to the test. But at least now, we've got the theory down on virtual paper to test. To tweak, improve, and try again if it fails. To lather, rinse, and repeat if it pans out. And remember: If it does work out, you've got Northrop to thank for playing the role of guinea pig. Send Northrop a thank-you note for providing the backlog data that made this experiment possible.

For further Foolishness:

Fool contributor Rich Smith owns shares of American Science & Engineering, which is a Rule Breakers selection. The Motley Fool has a disclosure policy.