Once again, SunPower (NASDAQ:SPWR) crushed earnings estimates -- and once again the stock is dropping like a rock the following day. This is becoming a theme repeated quarter after quarter.

Short term, investors were disappointed in a ballpark 2014 earnings guidance of $1.00 per share, lower than Wall Street's $1.23 estimate. Long term, SunPower is laying the foundation for a diversified business with more capacity and lower costs. Here's what the company reported and what you should take away from the third quarter.  

The numbers
Revenue was up slightly from a year ago to $657.1 million, although we shouldn't get too worked up about revenue because it is volatile quarter-to-quarter depending on when projects are completed.

GAAP gross margin was 29.4% and earnings were $0.73 per share, but GAAP numbers include a one-time gain and will jump around based on projects. So, I'll focus on non-GAAP figures, which level out project earnings and pull out one-time items.

Non-GAAP gross margin was 19.1%, down slightly from 19.5% last quarter but still near the top of the industry. Earnings per share were $0.44, well above the $0.15 to $0.35 per share guidance range given for the quarter.

SunPower installed 16 MW in the residential market, which was down from 18 MW last quarter and falls well behind the growth of SolarCity (NASDAQ:SCTY). I'll cover more about the reason for the drop below.

The big news was an announced 350 MW capacity expansion in the Philippines, due to begin production in early 2015 with full ramp up by the end of the year. Due to efficiency improvements, capacity will be up about 10% next year to 1.3 GW and with the new expansion the company is targeting around 1.8 GW of capacity in 2016.

Behind the numbers
One of the important markets to watch this quarter was Japan and it was strong once again. About 26% of the 313 MW produced went to Japan and APAC had the strongest margin of any region at 22.4%. High efficiency modules are perfect for the land constraints of Japan and I expect SunPower to continue to be a strong player there.

Another key I pointed out before earnings was utilization. SunPower's plants were running full steam last quarter and the 313 MW produced is actually slightly ahead of stated 1.2 GW capacity. One of the challenges SunPower faced was allocating where panels would go since it was constrained during the quarter. As I stated above, the residential market shrunk slightly and that's one of the reasons it did. There simply weren't enough panels to go around. Obviously, SunPower wants to take advantage of the high margin Japanese market while it can, it has to build contracted utility scale projects in the U.S., and it wants to keep a foothold in Europe, so SunPower won't grow into residential leases as quickly as SolarCity has even if it had the demand to do so.

Another market taking away capacity from residential solar is commercial installations. These are rooftop or small ground mounted installations that companies like Apple, Macy's, and others are installing in big numbers and SunPower has built a $1 billion backlog of these commercial projects. CEO Tom Werner said these projects will be installed over two to three years, and considering that the backlog will use about a quarter of the company's annual production this is a huge market. The advantage of commercial solar for SunPower is that acquisition costs are lower and the installations are larger. SolarCity has spent hundreds of millions of dollars acquiring companies that sell solar and keep its installers busy. SunPower can win a $60 million project (about the size of Apple's 20 MW Nevada project) from a single customer in the commercial market.

The really big news was the 350 MW in capacity expansion. SunPower is building this for its new X-Series modules, which are currently 21.5% efficient. These high-end modules currently account for about 10% of the company's production but management wants to bump that up to 30% or more in the future. While it wasn't announced, we can also expect the new manufacturing plant to produce an even higher efficiency product at lower costs. I wouldn't be shocked to see SunPower making a 23.5% efficient module by the time production ramps up in late 2015.

While the market is concerned about guidance today, it's this capacity expansion that's going to hold back the bottom line near term. There are operating costs that will go into building the new facility and equipment testing is expensive as well. Those costs come out of the bottom line next year with the payoff coming in 2015 and beyond. Also, keep in mind that dilution from a convertible offering and rising stock price will pull earnings per share down next year even if net income doesn't drop. Also, management has low-balled guidance for over a year. My guess is that actual earnings for next year will be between $1.50 to $2.00, just based on recent performance and guidance figures. At the least, $1 is the low end of what investors should expect.

The big picture
SunPower is laying the groundwork for becoming a major solar player in every country in any size installation. It's the only company that has that global reach and has share in residential, commercial, and utility scale projects. So, while it may not be growing as quickly as SolarCity in residential solar it's keeping a foot in the market with the intention of growing in the future when it has more capacity.

Last quarter's results were outstanding, and even though guidance was disappointing I think the future is bright. SunPower is the long play in solar, and though it may not grow profits in 2014 it's preparing for growth in 2015 and beyond. I still think this is not only the safest but also the best bet in solar right now. Blowout numbers like last night will eventually be reflected in the stock price for investors willing to make the long play in solar.

Fool contributor Travis Hoium manages an account that owns shares of SunPower and personally owns shares and has the following options: long January 2015 $5 calls on SunPower, long January 2015 $7 calls on SunPower, long January 2015 $15 calls on SunPower, long January 2015 $25 calls on SunPower, and long January 2015 $40 calls on SunPower. The Motley Fool recommends SolarCity. The Motley Fool owns shares of SolarCity. Try any of our Foolish newsletter services free for 30 days. We Fools may not all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. The Motley Fool has a disclosure policy.