by Julian Spector
December 20, 2019

The conventional approach to commercial energy storage didn't collapse or disappear in 2019, but its practitioners altered their strategies in notable ways.

Changes were already afoot. But this year began with the commercial business model — customer-sited demand reduction, grid services, value stacking — seemingly alive and well, and ended with it largely sidelined by its most ardent advocates.

Advanced Microgrid Solutions, Green Charge Networks and Stem, the early leaders in the space, haven’t gone anywhere. But all three have shifted from the early vision that they pitched to venture capital firms to raise a collective total of hundreds of millions of dollars. 

AMS conceived of a pivot to software in 2017, after investors balked at the startup's approach to develop capital-intensive, multiyear storage deployments for utility contracts. This year AMS confirmed its first customers for its newly developed software, which uses AI to crunch a multitude of variables to dispatch power plants, including energy storage, in wholesale markets for maximum return. It contracted to dispatch several thousand megawatts in Australia’s competitive energy market, and was planning to enter California wholesale markets as of November.

Green Charge made its name installing batteries at convenience stores, schools and small businesses to help them save money on demand charges. It achieved sufficient revenue to attract a buyer in Engie, the French utility actively embracing grid edge technologies in the U.S. market. This year, the rebranded Engie Storage turned its attention from California commercial customers to front-of-the-meter solar plants in Massachusetts, which were newly incentivized to add batteries.

And Stem, after years of building a business on value-stacking behind the meter, this year slashed its regulatory team and development staff and reoriented around recruiting installer partners to deploy Stem's storage software and expertise.

When all the pioneers of an emerging business model start doing things differently, it's worth asking why. In aggregate, these changes call into question the value of commercial energy storage as practiced in the early years of the market. But 2019 also saw the completion of projects that test out new approaches to commercial storage and address the weaknesses of the old ways.

Where is the growth?

Commercial storage had its best year ever in 2019, with 118 megawatts/293 megawatt-hours deployed, according to Wood Mackenzie. But it’s clearly the smallest market segment, trailing both the utility-scale sector and the formerly tiny residential sector.

And commercial storage has never had a clean growth story to tell from quarter to quarter. After a record Q1, installations dropped roughly in half in Q2, and then dipped slightly lower in Q3, in keeping with the industry's history of ups and downs.

The bigger red flag in the data has to do with geography. After years of chipping away at this product, companies have only achieved a sizable market presence in California, where the Self-Generation Incentive Program subsidizes the cost of batteries, and a smattering of early utility procurements gave the early adopters firm revenue to support the variable revenue of demand management contracts.

That SGIP funding could soon dry up, however. California regulators will soon vote on a proposal to direct $513 million from SGIP to low-income and medically vulnerable families in the wildfire zones — and slash the share directed at commercial batteries. 

Outside of California’s favorable environment, commercial storage has yet to achieve the kind of growth that makes VCs happy. But WoodMac’s outlook projects that New York and Massachusetts will likely be joining the party over the next five years, with small uptake in Arizona, New Jersey and a smattering of other states.

When evaluating commercial storage, it's important to remember that the market is still young, said WoodMac analyst Brett Simon, who tracks behind-the-meter storage. Much of the value proposition hinges on distributed energy resources being allowed to compete in wholesale markets, he added. But that vision has yet to become a market reality in most of the country.

To be sure, after years of deliberation, the Federal Energy Regulatory Commission's Order 841 is starting to create energy storage market pathways in the regional grids that FERC oversees, he said. "FERC Order 841 has now been resolved and will start being implemented, which should create some upside," Simon said. That regulatory change could create vast new markets as the cost of batteries keeps coming down.

But if wholesale market access has lagged behind the hopes of the storage industry, companies have found refuge in other policies, like Massachusetts' SMART incentive, which pays extra for solar with batteries attached, or the federal Investment Tax Credit, which chopped 30 percent of the cost of storage that charges from solar. 

Indeed, there's a case to be made that the pivots and shifts in focus from commercial providers represent a sign of health for the industry. "There's still room for innovation and growth — market penetration is still super low, and there are still many customers with high demand charges," Simon said.

For one thing, nobody has gone bankrupt over this strategic gamble — yet. Green Charge cashed out with Engie. AMS raised a boatload of money for its software and signed up paying customers this year. Stem laid off some employees but has revenue flowing in from its 600+ operating assets, and has closed at least a few deals as software supplier to other people’s projects. 

If anything died, it was VCs' hopes of rapid growth to deliver a healthy multiple. What they got was a few years of state-supported projects in California, a small number of projects elsewhere, and a transition from project development lingo to software platform jargon. At least they switched to a plausible alternative, rather than, say, cell tower backup in remote island nations. 

Newer models of commercial storage made headway this year, proving that alternatives exist. If the new approaches had anything in common, it was simplification and scale.

Get that customer out of here

The classic value-stacking strategy needed several things to work in concert.

The operator had to accurately predict when the host customer would hit a consumption peak, then dispatch the batteries in time to clip it. That action had to balance with other obligations for grid services or wholesale market participation. Stepping back in the sales cycle, the company must spend months, if not years, wooing businesses that have an addressable peak pattern, and then have the patience to get comfortable with a new technology.

This year witnessed the arrival of a new approach to customer-sited storage: Cut out the customer. More specifically, engage them as a landlord, pay them money to use their space, but don’t involve them in battery operations in any way.

Developer GI Energy pioneered this structure in New York City. The company, now owned by Shell, won a demo project with Con Edison in 2017 and got systems up and running this year. The utility liked it because it makes it easier to target locations that maximize grid benefit, rather than hoping eligible demand-management customers roughly line up with grid stresses.

As Con Ed’s Alison Kling told me at the time, this approach drastically simplifies the customer outreach.

"Instead of talking to customers about demand charges, GI can talk to them about a lease payment," she said. "They know what's coming, and that's the language they work in."

Following that breakthrough, Enel X confirmed this month that it completed the largest lithium-ion battery in New York City using this same “customer-sited, front-of-meter” arrangement. The company was a little cagey on where all the revenue comes from, but it’s definitely cutting a lease payment to a real estate company to make use of square footage at a shopping mall.

Enel X Vice President Philip Martin declined the opportunity to deliver a full-throated critique of the old VC-backed model, instead describing the new structure as “another arrow that we can put in our quiver.” 

But the company’s press release announcing the project praised it for “removing the complexities of energy management and tenant participation from the structure.” It’s hard not to read that as an indictment of former industry practices.

Even Stem's move to deliver storage through channel partners reflects a desire to streamline the sales process.

"In aggregate, we just want to do more partner work, because from a [customer-acquisition cost] standpoint, it's obviously lower, and that's an important metric for us as a business," CEO John Carrington told me in an interview this summer.

Stem still stands by the value it can deliver to customers; it hasn't tried to turn them into mere lessors of land. But if a local solar installer who's in the neighborhood wants to get potential customers jazzed about storage, that's easier for Stem than pitching directly to every end customer.

Avoiding "diseconomies of small scale"

Smaller batteries in smaller businesses kicked off the commercial storage market, but the action has shifted to bigger batteries for large industrials.

Small batteries, naturally, have less capacity to provide valuable services than do big ones. Distributed storage companies compensate for this by playing a numbers game: If they can simply deploy vast amounts of batteries and coordinate a network, they can add up to a power plant’s worth of capacity, while keeping their customers happy.

Residential installers like Sunrun and Swell Energy are banking on this theory as they sign up customers for thousand-home aggregated capacity contracts. But they have the benefit of closing deals face-to-face with individual homeowners (and even then, it should be noted, we haven’t seen these fleets come to fruition yet — they’re still in the early stages).

Commercial storage companies have to convince professional energy managers and navigate corporate procurement channels. The process often takes one to two years. The upshot is that C&I storage doesn’t install many systems — just 431 over the last four quarters on file, according to WoodMac. The ones they do have been on the small side, to keep projects from costing too much and to match small customer loads.

These dynamics produce what one storage industry veteran described to me as “diseconomies of small scale.” The compromises made to place a grid asset in a business property yield a less profitable grid asset than if you built one for that purpose alone.

"A lot of it comes down to effort versus return," said Simon, the analyst. "Why spend 12 to 18 months on a 10-megawatt-hour project when, for the same time investment, you could snag a 50-megawatt-hour project?"

The easy fix to this dilemma is build bigger, so the investment in customer-acquisition costs produces an asset with lower unit costs and bigger margins. That’s exactly what the C&I storage pioneers did this year, implicitly confirming the critique.

Engie Storage stacked a different set of values than its old solo battery-in-schools template. The company switched its focus to front-of-the-meter solar plus storage projects in Massachusetts. These systems earn a federal Investment Tax Credit and the state’s SMART incentive with a storage adder. On top of that, Engie now pays developers for the right to dispatch batteries into the wholesale markets, using the global energy conglomerate's talent for energy trading and risk analysis.

Stem also crossed to the front of the meter and began partnering with solar developers to add storage, rather than developing its own standalone storage. This summer, it announced a deal with private equity firm Syncarpha Capital, also in Massachusetts (That same firm is working with Engie, perhaps comparing and contrasting two similar services). In November, Stem executed a master supply agreement to equip NEC Energy Solutions projects with Stem’s economic dispatch software, setting the stage for more utility-scale projects to come.

To the extent that Stem is still developing its own projects, it’s particularly excited about the Ontario market, where the "Global Adjustment" charge justifies large batteries to help industrial facilities avoid hefty bill charges. These projects can also layer in power quality and even short-term backup services. Stem raised a CAD $200 million financing facility from the Ontario Teachers’ Pension Plan in the summer of 2018 to deploy batteries in the province. 

At the time, Carrington said he would put that capital to work in the coming 12 months; that interval has passed and the company hasn’t said much about what it’s built there. But Carrington confirmed in an interview this summer that Stem had installed its largest battery system for an undisclosed customer there and that other installations were under way. Stem's website indicates it installed a 2.1-megawatt/3.4-megawatt-hour system for manufacturer Solvay.

Others have delivered, too. Convergent Energy + Power built the largest behind-the-meter battery in North America: 10 megawatts, to support an industrial facility in Ontario. Last month, the company tied its record with a new system for Shell. Along the way, private equity firm Energy Capital Partners liked what it saw in those projects and acquired Convergent, giving it a direct source of capital for future projects.