Accelevation

Moat sources are real but not impregnable. Switching costs arise when Accelevation is designed into hyperscale programs and modular white‑space packages, where spec‑in, site standards and execution familiarity discourage vendor churn.

Vertical integration and 100% U.S. production across multiple facilities enhances speed, lead‑time reliability and customization, conferring a localized cost and service advantage vs. import‑reliant competitors.

Brand equity with leading hyperscalers adds intangible value, and the company’s integrated install capability tightens customer lock‑in on multi‑phase campuses. That said, these are execution‑dependent advantages in a market where large incumbents (ABB, Eaton, Schneider, Legrand, Vertiv) possess scale, channel breadth and deep product catalogs.

There is little to no network effect, and commodity inputs and project pricing pressure can erode advantage. Overall, we view a moderate, execution‑based moat driven by spec‑in and vertical integration rather than structural network dominance.

Evidence of pricing power is mixed. The company’s ability to bundle engineering, factory‑built modules and installation creates value vs. fragmented multi‑vendor builds, supporting premium pricing on schedule‑critical AI deployments. TTM Adjusted EBITDA margin of roughly 18% on ~$727m revenue signals reasonable unit economics.

Yet competition in power distribution and containment remains intense and substitution costs, while real, are not insurmountable for hyperscalers. Input cost volatility (steel, copper) and large‑account negotiations limit unilateral price moves.

We see some latent pricing power in next‑gen, high‑amp power systems and modular platforms as rack densities rise, but we do not ascribe monopoly‑like economics.

Backlog of about $1.11 billion as of 2Q26 and programmatic hyperscale builds provide near‑term revenue visibility. However, the revenue model blends product with field services and remains exposed to program timing, customer capex cycles and order lumpiness.

The S‑1 highlights dependence on a concentrated set of hyperscale and colocation customers and irregular ordering patterns, which can challenge quarter‑to‑quarter predictability. In our framework, this is more predictable than many construction firms but below toll‑booth‑like businesses.

Leverage is the main weakness. Pro forma for the IPO, long‑term debt declines from about $642m to roughly $464m, with cash near $19m at June 30, 2026. This implies net debt of ~ $451m and limited cash cushion. A newly recorded TRA liability around $82m sits above common and can siphon future cash tax benefits to pre‑IPO owners.

In a 5%+ risk‑free environment, floating‑rate exposure magnifies downside in a demand pause. The business is cash‑generative (TTM FCF ~$50m) but not yet fortified for severe cycles given concentration and capital needs.

Strategy centers on organic capacity adds and program wins, complemented by M&A (e.g., Instor) that broadened from containment into installation and white‑space services under prior ownership, and a 2025 change of control to Olympus Partners.

While integration created a stronger offering and supported rapid growth, the Up‑C structure, TRA, and a large 40m‑share S‑8 for equity incentives heighten dilution risk. IPO proceeds primarily reduced debt rather than funding outsized organic bets, which is sensible but leaves leverage meaningful.

We do not yet have a track record of disciplined buybacks or clearly accretive, measured M&A as a public company.

Founder‑CEO Michael Rubiera remains at the helm, and the post‑IPO board includes experienced operators and Olympus representatives. Founder continuity is a positive for customer relationships and engineering culture.

That said, control resides with private‑equity affiliates and Class B holders, and the Up‑C with TRA reduces alignment of public shareholders with total free cash flows. We will watch equity comp practices and capital deployment discipline closely in the first 6–12 quarters as a public company.