Behind-the-Meter Storage:
When the Math Works
And When It Doesn't.

01The Setup

Behind-the-meter storage has three revenue streams: demand charge reduction (shaving peaks to reduce capacity-billed costs), time-of-use arbitrage (charging when energy is cheap, discharging when expensive), and ancillary services / DR (selling flexibility back to the grid).

Capital costs have come down — current installed cost for 2-hour systems above 1 MWh runs $300-400/kWh — but operational complexity is real. The 30% federal Investment Tax Credit (ITC) materially improves project economics, and state incentives stack in specific states (CA, NY, MA).

The economics depend critically on local tariff structure, market access, and load profile. A site with high demand charges, DR market access, and backup-need value gets very different economics than a site with flat-rate energy and no backup requirement.

02The Data

The 'where it works' pattern: high demand charges (NYC, parts of California), DR-accessible markets (NYISO Zone J/K, ERCOT), sites where backup has independent value (hospitals, data centers, food storage). Two of three conditions usually clears the math; three of three makes it obvious.

The 'where it doesn't' pattern: flat-rate energy, low demand charges, no DR access, no backup need. Most commercial sites in low-cost utility territories with no time-varying rates simply don't pencil — no amount of slide deck enthusiasm changes that.

Federal ITC plus state incentives stack to materially improve project economics in specific states. California's SGIP, New York's NY-Sun and behind-the-meter storage incentives, Massachusetts SMART program — all change the math meaningfully for sites in those states.

03The Implication

For the right customer profile, behind-the-meter storage now offers 5-7 year payback — competitive with operational alternatives. The challenge is identifying which customer profiles fit and which don't, before the customer hears the pitch and asks for it everywhere.

Sustainability narrative: storage doesn't reduce emissions directly, but it enables higher on-site solar penetration and load flexibility that does. Bundled with solar, it's a real decarbonization story — and the bundled economics are often better than either alone.

04The Recommendation

1. Audit candidate sites by demand charge level, TOU spread availability, DR market access, and backup value. Two of four is the screening threshold; three or more is the obvious yes.
2. Build the stack model: which revenue streams, with what assumptions, with what state/federal incentive treatment. Don't sell on one revenue stream and forget to model the rest.
3. Don't recommend storage as a general-purpose solution. Recommend it for the specific customers where the stack supports it. The customers who hear 'storage everywhere' are right to be skeptical.
4. Watch utility tariff filings — TOU structure changes can shift economics meaningfully. Sites that pencil today may pencil less in 18 months if rates restructure.

Storage is one of the technologies where the slide deck math and the actual project math have diverged for a decade. The gap is finally closing in specific markets. The advisor's job is to know which customers and which sites fit the math, and which don't. Most don't — yet. But the list of 'don't' shrinks every year.