Eos Energy Enterprises pre-announced second-quarter 2026 results on July 15 that materially reset the frame on US long-duration storage. Preliminary revenue of $68 to $69 million is a company record, more than three times prior-year shipments by volume, and pushes first-half 2026 revenue past the company’s full-year 2025 total. Backlog reached approximately $807 million as of June 30, up roughly 25 percent from the prior quarter. Cash including restricted cash was about $364 million; customer collections of about $78 million exceeded quarterly revenue, which matters for a company that has burned cash through the ramp.
Two things are worth pulling out.
First, this is a non-lithium chemistry (zinc-halide aqueous) hitting scale on US soil. Most of the domestic-content storage conversation has been about lithium iron phosphate cells and where the anode/cathode active materials come from. Eos sidesteps that supply chain entirely: its cells contain no lithium, no cobalt, no nickel, and no critical minerals covered by the foreign-entity-of-concern rules under Section 45X. Battery Line 2 at Thorn Hill in Turtle Creek, Pennsylvania started commercial production this quarter, and the company reiterated a 4 GWh annual run-rate target by year-end. For projects trying to qualify for Section 48E investment credits with adjusted-percentage domestic content, an American zinc-halide option changes the pricing dynamic on longer-duration bids.
Second, the backlog composition tells the demand story better than the revenue print does. The 2 GWh firm capacity reservation agreement with Frontier Power USA, the developer joint venture Eos formed with Cerberus earlier this year, produced its first purchase order last month: 100 MW / 400 MWh for the Redbird project. That is the mechanism by which the backlog turns into revenue on a predictable cadence rather than deal-by-deal. It also aligns with where grid-scale demand is going: durations of 4 to 8 hours to firm renewables and shoulder data-center diurnal load, not the 1 to 2 hour arbitrage duty that dominated the first wave of US battery build-out.
Where to be careful. Eos is a single-supplier data point, not a market. The stack of announcements around US battery manufacturing this year (Honda-LG’s Ohio pivot to stationary cells, T1 Energy and Kore Power sodium-ion plans, and now Eos’s second production line) rhymes but the aggregate US domestic non-Chinese cell supply is still small relative to the 35 GW / 70 GWh of interconnected 2026 US BESS deployment that SEIA is now guiding to. The bottleneck for the next 24 months is refining and cell manufacturing capacity that can clear FEOC and adjusted-percentage domestic content thresholds; Eos moves the needle in one narrow lane of that fight.
Watch: full Q2 results on the previously scheduled conference call, particularly gross margin at the new run-rate and the composition of the incremental $161 million of backlog added this quarter. If Frontier’s purchase-order cadence continues at the current pace, revenue guidance for the second half of 2026 becomes the more interesting number than the Q2 print itself.