Blog Post

To reduce the cost of solar, look to everything but the hardware

“Soft costs.” That term was repeated on stage, in workshops and in the hallways of the U.S. government-organized SunShot Summit this week in Anaheim, California. More than 800 of the world’s most dedicated researchers, business execs and policy makers attend the event to focus on how to reduce the costs of solar in the U.S. dramatically by 2020.

Disneyland might be just a few blocks from the Summit, but these folks are hunkered down in an over-air-conditioned hotel talking about things like the limits of solar cell band gap, robotic panel trackers and the nuances of power grid integration. Their version of riding Splash Mountain is a workshop on net metering.

Soft costs include everything that has to do with installing solar panels on rooftops (and large solar projects in the desert, for that matter) except for installing the actual solar hardware — things like permitting, financing, taxes, marketing and customer acquisition, labor and supply chain costs. When it comes to installing solar panels on the rooftops of homes across the U.S., soft costs made up a whopping 64 percent of the total cost of the system, according to a report out late last year from the National Renewable Energy Lab (relying on data from 2012).


That means there are a variety of ways to incrementally reduce the costs of solar rooftops in the short term without inventing new technologies or delivering breakthroughs in materials. It’s more about streamlining the processes, using data to make everything more efficient and building networks that reduce the time (and thus the money) needed to get solar systems installed.

In the SunShot Program — which is modeled on the Department of Energy’s early-stage high-risk ARPA-E grant program — reducing soft costs is just one of five areas that is the subject of funding and focus. But SunShot director Minh Le called it perhaps “the toughest job” out of the five areas, partly because it’s about “human behavior.”

For entrepreneurs and investors, though — particularly in Silicon Valley — the soft costs could provide the most opportunity for innovation. Material science breakthroughs take forever and are extremely difficult. Leveraging computer science and mimicking more efficient systems, on the other hand, seem like low-hanging fruit.

Apple's solar farm
Apple’s solar farm

Elaine Ulrich runs SunShot’s soft cost program (officially called the Balance of Systems program). She was a former senior aide to Gabrielle Giffords (including during the time period of that tragic event), and has a Ph.D in optical science from University of Arizona.

During a talk at the Summit on Tuesday, Ulrich said her group is heads-down investing time and money in “data, tools and skills” to reduce solar soft costs. “Data is key,” she emphasized, “but it’s about connecting data with people.”

Each of the five SunShot departments launches a variety of “solicitations” — calls for university labs, national labs and companies to submit their projects and ideas to receive funding or other incentives. The Soft Costs division has worked on solicitations like a solar instructor training network, an incubator program and a project to use analytics to analyze solar case studies from conception to deployment.

On Tuesday the SunShot program announced a new contest around soft costs, called SunShot Catalyst, that offers between $500,000 and $1 million to groups that use “automation, algorithms, data and software,” to solve solar problems. A lot of startups out there are already doing this, so SunShot Catalyst shouldn’t have any trouble getting submissions. The SunShot Summit is also highlighting a solar web and app hackathon starting Wednesday night.

Soft costs around financing solar projects could be one of the most important areas for cost reduction. Neww financing business models for residential solar rooftops have already unleashed a market in the U.S. and made the innovators rich in the process. But financing large solar panel utility-scale projects — like the ones SunPower and First Solar build — are still more expensive than many other types of energy generation projects. The CTO of First Solar, Raffi Garabedian, said that he thinks bringing down the cost of financing utility-scale solar panel projects could be the most important result of the SunShot program.

In the long run, the solar industry will need to make solar panels themselves more efficient and cheaper, and breakthroughs with next-generation panels like CIGS will rovide some of those needs. But in the short term, to help meet the goals of the SunShot program — to get solar projects to the grid parity cost of 6 cents per kilowatt hour by 2020 — the program only has six years to reduce solar costs, and a big chunk of those reductions will come from the soft costs.

15 Responses to “To reduce the cost of solar, look to everything but the hardware”

  1. The sole procedure delaying my switch is confronting the paperwork for a loan. Essentially, we have to apply for a first mortgage to acquire a first class system. Daunting, even though we have the credit to qualify.

    We have the space – and plan to do a ground-based system. Sooner or later more counties like ours will make this first choice where practical because of obstacles presented to firefighters with roof-based installations.

  2. Matthew Wright

    Cutting electrician’s out of the picture is one of the main soft-cost reductions that we require (especially in Australia where I live).

    The answer is to return to 48V systems (80VOC) 2 panels in series and multiple strings. What we require for this is the inverter manufacturers to put together inverters that have a MPP of 60V-160V (2 or 3 panels in series) and 4 of these MPPTs per inverter with a 30Amp max so you can put signfiicant number of panels on the inverter.

    This eliminates electricians from the rooftop wiring game and reduces the cost of labour considerably.

  3. Reading the NREL report is depressing. Most of these “soft costs” are laughably bogus.

    If I want a new electric panel installed, I can order the parts from Home Depot and hire an electrician to do the work who I more or less pay by the hour. If I don’t like a monolithic quote one guy gives me, I can get three.

    Solar is nonsense by comparison. I can buy panels online, so why am I paying “supply chain costs”? Why am I paying “customer acquisition costs”? Why can’t I just get someone to do the installation who I pay by the hour? It sure seems that if we had thousands more small businesses trained to do installation, soft costs would fall by more than half and yet all those folks would still be making 100 cents on the dollar for their labor, the one part of the cost equation that is legitimate.

    “For residential systems, the greatest soft costs were supply chain costs ($0.61/watt), installation labor ($0.55/W), customer acquisition ($0.48/W), and indirect corporate costs ($0.47/W), such as maintaining office management and accounting functions.”

    That’s close to $2/watt in basically fake costs. Part of that is reasonable for customers that aren’t inbound requesting systems; next to none of is reasonable for people making their own requests and procuring their own parts. Maybe we need Ikea here like they have it in the U.K. That could take that $2 down to probably 20 cents overnight. With panels at 75 cents, you’d be under $2 watt installed and solar would essentially make sense in about 2/3 of the CONUS.

    As for permitting, it’s completely unclear why there isn’t some sort of national standard. The federal government has done a good job in the past of figuring out how to supersede really unimportant state rules for the greater good. That said, how has California failed to make a rule that limits permit costs for solar to $100, period? That seems like more than enough.

  4. DIY is difficult if not imposible for net metering because it involves the utility company. You need to sign contract and get their approval. Second, roof top installation need to get Department of Building involved and approved. The solar panel installer are young guys that can carry the big panels up the roof and do the installation. The job is for the professionals. What I hope is that the process can be streamlined and the knowledge that the industry learn can help us to publish some standard procedure and hardware that will further reduce the cost. That is what the governement incentive and rebate meant to do and I think it will succeed to some degree. I am not sure it can reach the 6 cents per KW objective.
    Alternative energy has a long way to go. It should still be our goal.

  5. Sammy Solar Park

    DIY has the best ROi for solar, and if you use Solar Clam-P mounting hardware, that will also reduce your overall cost. Especially when it comes to any maintenance issues down the line.

  6. Kathleen

    I live in New Jersey and the Utilities do not want more people to go solar. It seems the companies are losing money, so they don’t want to pay the people for the excess electricity being put into the Grid. What do we need to do to stop this problem before it gets any
    worse ?

  7. Hans Berg

    This critical issue seems like it is largely being over-thought at the federal level, at least from my perspective in the residential solar space. I have dealt with many AHJ’s in multiple states and the variation in permitting processes and procedures is maddening.

    What is so difficult about creating some robust (but simple of course) national standard permitting procedures and pushing them down to the local levels? Solar is simple; that needs to be the message from above. So many local jurisdictions are trying to be too smart and without federal standards or guidance are using the mystery to overcomplicate and overcharge. This talk of “data, tools, and skills” plus “automation, algorithms, data, and software” sounds highly bureaucratic and counterproductive. Perhaps there is too much money available to solve what I see as a fairly simple problem.

    Also, most certainly a “solar training network” is not our problem right now (unless it is for the AHJ’s I suppose). Simplify the process, let’s get 5% of roofs covered, and the training opportunities will explode organically with the tipping point.

    Again, my comments are based on the residential space but here solar is not complicated. Deploy, deploy, deploy….

  8. Joshua Posamentier

    While reducing soft costs are great, there are small changes on the hw side that cascade into the soft cost picture. You don’t see fslr panels on rooftops, right? That’s cheaper hardware but doesn’t yield a cheaper solution in all scenarios.

    • Spec9

      First Solar panels are thin-film panels. Since roof-top space is very valuable, residential installers tend to use more efficient silicon panels to maximize returns.

  9. Spec9

    You want to install solar PV for cheap? Do it yourself. If you are a DIYer that can install a new 240V outlet and install an attic roof vent then you have all the skills you need to install a microinverter based rooftop solar PV system. If you install yourself, it is DIRT CHEAP and you’ll get free electricity for the next 25+ years.

  10. Albert Hartman

    There’s an implied assumption that high cost is what’s holding up widescale solar adoption. For a growing demographic, that is no longer true. You can already see it California’s grid utility demand that now shows a dip in the middle of the day, owing to all the existing solar sources coming on line. Just like more expensive smartphones pushing aside cheaper dumbphones, when people decide the non-financial benefits outweigh the alternatives solar momentum will surge forward. There are non-monetary considerations like wanting less-polluting energy sites, wanting to use 21st century tech solutions, wanting to sever financial ties with old corporate players engaged in malfeasance. Arguably the day is quickly coming when financial considerations will fade as the largest concern.