UC Davis power plant, built by SunPower. Image by the author.

The solar industry is a fickle business for investors, largely because of the rise and fall of businesses and market segments year after year. Investors have seen high-profile panel companies go bankrupt, developers go out of business, and more recently residential solar companies drop like a rock.

To combat this volatility, strong solar companies will lean on diversification wherever they can find it. In the case of SunPower (SPWR -9.82%), diversification around the world and in all segments of the market is the company's biggest strength and what makes it most confusing to investors. And this week's third-generation Oasis power plant launch showed how that dynamic plays out for the company.

Two years ago, SunPower was planning on leaning on LCPV for growth. Image source: SunPower 2014 analyst day presentation.

Changing course again

Two years ago, I attended SunPower's analyst day in New York and had the chance to sit down with CEO Tom Werner for a lengthy interview. SunPower was beginning to recover from the depths of the solar market in 2012 and had recently introduced some new products that management thought would take the company to the next level. Key to the plan to triple production between 2014 and 2019 was a concentrated solar product, or LCPV, called C7 for the power plant business. You can see in the chart from that analyst day that by 2016 500 MW of LCPV was expected to be installed with 1,500 MW on the horizon in 2019.

Less than two years later, LCPV is barely mentioned at SunPower. The product still exists, but SunPower now sees its P-Series module, which is slightly more efficient than a commodity solar panel and only slightly more expensive, and system design capabilities as their differentiators in utility-scale solar. I'll explain why P-Series makes sense in a moment, but you can see that the strategy can shift quickly.

Meanwhile, the high-efficiency solar panels that have always been SunPower's differentiation are being shifted away from utility-scale power plants to residential and commercial solar. This makes sense given their advantages in space constrained markets and it leverages SunPower's diverse customer base, but that's a shift from high-efficiency cells driving LCPV's performance two years ago.

SunPower thinks that, on the utility side of the business, moving away from LCPV and high-efficiency panels to a lower price panel and more services oriented offering will bring in new customers and allow it to play in new markets. And that's where the third-generation Oasis product comes into play. 

The drone SunPower will use to plan power plants. Image by the author.

A new strategy emerges

Oasis is SunPower's tracker design that has been in the field for six years. It formerly used high-efficiency SunPower panels and was meant primarily for internal project development. The third generation of Oasis design, announced this week, moves away from a specific panel and self-development, but it tries to optimize everything from site design to operations and maintenance. And those services are what SunPower intends to sell to developers. 

The drone shown on the left is really where Oasis begins. It flies above a site and using SunPower's GEO software generates 10^60 potential power plant designs, considering everything from elevation changes to different solar panel options. In a relatively short amount of time, SunPower can figure out what its costs and return on investment would be for a given site and what it may have to charge for energy. That's a revolutionary development by itself.

The drone data then combines with the Oasis design that uses 50% fewer parts and can pack 60% more energy production in the same space than previous versions to create a complete power plant system. SunPower does this by making the arrays wider and unlinking them from each other, allowing more arrays to be packed into the same amount of space as before. 

What's changing in SunPower's strategy is how it's using GEO and Oasis to make money. The company has been moving toward a self-development model for the last three years, but Oasis will be sold as a service and components, particularly in international markets. A customer can even have SunPower come in and do a site design for an Oasis system that uses a third-party solar panel if it wants to. 

The component strategy is something First Solar (FSLR -6.16%) is shifting to as well, but SunPower has taken a more holistic approach. And First Solar has taken focus on less of the design process with its Series 5 product, which won't even launch until 2018. That could be a differentiator for SunPower going forward. 

It's likely SunPower will use a multipronged strategy for Oasis sales, using different strategies in different regions. The U.S. may still be mostly self-developed and places like Africa could see only component sales. What this could do is give the company sales in emerging markets without the risk of developing projects itself. If you look at the recent bids in Abu Dhabi of 2.4 cents per kWh or Chile for 2.9 cents per kWh, the return on building a solar plant is low for developers. SunPower may not want to take that risk in every case, but if it can still provide a cost-effective component solution for a developer that will take on the risk, it will be a win for the company.  

The new side of SunPower's power plant business

SunPower thinks a more holistic look at development offering will result in lower costs long term, which it will use in its own developments as well as sell to others. The drone capabilities and Oasis design, in particular, could be products that developers will find worthwhile and SunPower can scale without building a new manufacturing plant to provide the services. In that sense, it's a low-risk, high-potential offering to the market.

What investors should watch for is how quickly SunPower can scale these component sales and what margins will look like. There's a lot of potential, but a lot of unknowns as well. And considering how quickly the solar industry can change, it's no guarantee Oasis will be a hit with third parties.