Shoals Technologies is the boring solar stock with a wild setup

Every solar farm you have ever seen in a photograph is a lie of omission. The camera shows you the panels, endless rows of blue-black glass tilted at the sun, marching across a field in west Texas or the Atacama or some reclaimed lignite pit in Saxony. What the camera does not show you is the copper. The connectors. The junction boxes. The combiner boxes. The kilometres of cable trenched beneath the array, splicing 20,000 individual modules into strings, strings into circuits, circuits into inverters, inverters into the substation, substation into the grid.

That invisible layer has a name in the industry. It is called electrical balance of system, or EBOS, and it is the thing that decides whether a solar project comes in on time and on budget or turns into a lawsuit. It is also, unglamorously, where a huge chunk of a solar farm's construction cost actually lives. Panels get cheaper every year. Chinese oversupply has crushed module prices to the point that a watt of silicon costs less than a stick of gum. Inverters have become a commodity. Racking is a commodity. Land is land.

EBOS is the piece that has not commoditised. And in 2026, it is the piece the market cannot get enough of.

Here is why. American utility-scale solar has entered a phase most people covering the sector do not fully appreciate. The Inflation Reduction Act tax credits, which the current administration spent 2025 trying to unwind, survived in a mutilated but workable form after the December compromise. Domestic content bonuses are still in play. What that has done is force developers to source American-made EBOS if they want the full credit stack. Chinese wire harnesses that used to arrive in shipping containers now come with a 60% duty attached. American developers have exactly two choices. Pay the duty. Or buy domestic.

Meanwhile, the demand curve for utility solar in the US has gone parabolic for a reason nobody was pricing in 18 months ago. Data centre load. Hyperscaler power purchase agreements. Meta signing 12 gigawatts. Amazon signing 15. Google contracting for anything with a fence around it. Microsoft resurrecting Three Mile Island alongside its solar sourcing. When Sundar Pichai says Google needs to double its energy footprint by 2030, that footprint is not going to be filled by nuclear alone. Nuclear takes a decade to build. Solar takes 18 months. The math is not complicated.

So the utility solar order book in the US for 2026 to 2028 is now the biggest it has ever been. And every single one of those projects needs miles of American-made connector cable, combiner boxes, and the plug-and-play harness systems that let a construction crew wire up an array in weeks instead of months.

This is where the story gets interesting for a specific kind of company. The panel makers, First Solar included, are huge, well covered, and priced for perfection. The inverter guys are getting squeezed by Chinese entrants and margin compression. The EBOS layer has one dominant American player that reinvented how solar farms get wired, holds the patents on the plug-and-play architecture that most large developers now specify by name, and has watched its stock get treated like a pandemic-era leftover for the last two years while its addressable market tripled beneath it without anyone noticing.

The company was founded in 1996 in Portland, Tennessee, by a guy named Dean Solon who spent a decade wiring solar farms the old way, one crimp at a time, before he got sick of it and designed a better system in his garage. He sold pieces of the business over the years, took it public in 2021 at $25 a share, watched it rip to $34 and then bleed all the way down to single digits as the market decided EBOS was boring and residential solar was over and Chinese competition would eat everything.

The market was half right. Residential solar is a wasteland. Chinese competition is brutal. But EBOS for utility-scale, made in America, with the domestic content bonus attached, is a completely different animal. And the company that dominates it is now trading at a fraction of what its normalised earnings power looks like when the utility solar backlog actually converts into revenue.

Which brings us to Portland, Tennessee.