Five years ago, First Solar (NASDAQ:FSLR) was the undisputed king of the solar industry. Since then, competition has eaten into its lead, and net income has fallen in half, all because its thin-film panels that once led the industry were suddenly more expensive and less efficient than competing products.

FSLR Revenue (TTM) Chart

FSLR Revenue (TTM) data by YCharts.

That may soon change if First Solar can hit its own research and development goals, and an announcement on Thursday suggests it may be doing just that.  

The road to 21% efficiency
On Thursday, First Solar announced a new world record for cadmium-telluride, or CdTe, solar cells with an efficiency of 21.5%, as confirmed by the National Renewable Energy Laboratory. This puts the company ahead of plan in increasing its cell efficiency, which it hoped to reach 22% by the end of this year, as you can see below.

Source: First Solar.

The hope is that this will eventually translate to higher efficiency modules. In Q3 2014, First Solar's panel fleet efficiency was just 14.2%, compared to 14%-16% efficiency for commodity silicon-based modules, and a record efficiency of 21.5% from SunPower's (NASDAQ:SPWR) X-Series module.

It may look like First Solar is a step away from the industry's lead, but remember that solar cell efficiency doesn't translate to module efficiency, which will be a couple of percentage points lower. First Solar is heading in the right direction, though.

It's also a big step taking a solar cell made in a laboratory and transitioning to manufacturing. You can see in the research roadmap above that even First Solar thinks it will take two or three years or more to take today's technology into a manufacturing setting.

Making a 21.5% research cell is progress, and it's a long way before that kind of product comes off the manufacturing floor, but it's key that First Solar keeps pushing efficiency higher. Its survival is at stake, here.

Building large solar systems like this one is First Solar's main business. Source: First Solar.

Why First Solar has to get this right
The fact of the matter is that if First Solar doesn't make significant improvements in efficiency, it will be another cautionary tale in solar. Silicon-based manufacturers like Trina Solar, Yingli Green Energy, and JinkoSolar are already producing more efficient panels than First Solar's for less money, which is why First Solar sells very few panels individually today. First Solar's business has transitioned into selling complete solar systems, and it's a world-class project developer, so it's been able to maintain profitability. But that can only last so long if efficiency doesn't improve.

The good news is that First Solar's efficiency is scheduled to grow far faster than the efficiency of silicon based solar modules, so if it can execute, it could both grow and increase margins. Progress on some of those milestones is a good sign for that strategy.

Another ace up the sleeve of First Solar could be the TetraSun based silicon solar panels it has started producing. The subsidiary First Solar acquired last year just began production at a 100 MW facility that First Solar hopes will eventually make cells at 21% efficiency with low costs. Whether or not it can remains to be seen.

The bull case for First Solar relies on R&D
Any investor who thinks First Solar stock can rise in the future has to be betting on the company's R&D plans becoming a success. Without a better product, the company will slowly be passed by competitors, and it'll be nothing more than another solar project developer. For now, it appears the R&D investment is heading in the right direction, and that's a positive sign for the future.