Primoris: A 64% Drawdown, a Record Backlog, and a $450 Million Hole Pushed Into 2027
Renewables went from $2 billion to $1.7 billion in six months; everything else is being repositioned as the recovery story
PRIM · Earnings Call · 2026-11-04
The tape is already brutal; the call had to argue uphill
Primoris walked into its November 4 call with its stock down 63.7% from the May 5, 2026 peak of $202.92 — a decline steep enough that the last leg is the ugliest one, with a down-10-week:-47% segment in the whole-history shape before a partial bounce. Over the last 90 days the shares are still off 13.2%. This is not a name where a good quarter is a surprise; it's a name where the market has already rendered a verdict on 2026 and is now pricing 2027 expectations that management spent an hour defending. The quarter itself was genuinely better than the prior one. Revenue came in just under $2 billion, gross profit $204.7 million, gross margin 10.5%. But the last filed 10-Q (period end July 31) still freezes the trough in place, and it is ugly: gross profit of $82 million, an operating loss of $37 million, and a 4.9% gross margin. Guidance was reset down: adjusted EPS of $2.10 to $2.25 and adjusted EBITDA of $290 to $310 million for the year. The real news was where the miss came from. Renewables revenue is now expected at roughly $1.7 billion versus the $2 billion guide handed out in midyear, and, as Koti Vadlamudi put it, “approximately $450 million of revenue being deferred beyond 2026” — Koti Vadlamudi, President and Chief Executive Officer · 2026-11-04. Five of the six problem projects are complete, the last finishes construction this quarter. What is left is a 2027 argument.The words that quietly disappeared
The most telling shift isn't in the numbers, it's in the vocabulary. On the May call, Vadlamudi was explicit about the booking bridge: “we have verbal awards of $1.1 billion in the second half of this year, and another $2.8 billion that will sign” — Koti Vadlamudi, President and Chief Executive Officer · 2026-05-06. Six months later, verbal awards was one of Primoris' sharpest keyword decliners of the quarter — because management has stopped using it as a crutch. The language now is about what's signed: record total backlog just under $14.1 billion, MSA backlog up over $300 million sequentially, and a promise of "a strong fourth quarter of bookings." Storm work, the perennial Q3 upside kicker, also fell off the narrative, and Vadlamudi was candid that the quarter had "minimal storm restoration activity" — a reminder that the beat-risk in this business is weather, not strategy. What replaced it was consensus — a word that spiked in Primoris' keyword set this quarter precisely because analysts kept asking management to bless or beat the 2027 number. Vadlamudi obliged: “Given the trajectory, the positive trajectory in our backlog we feel confident in meeting or exceeding those expectations” — Koti Vadlamudi, President and Chief Executive Officer · 2026-11-04. That is a stronger claim than the guidance below it, and it rests almost entirely on bookings that haven't happened yet. When Philip Shen asked whether win rates had changed after six challenged projects, the answer was "no change" — but the proof point is a Q4 renewables booking number of roughly $800 million and a Q1 that's "maybe double."The thing growing underneath the hole
The genuinely new theme at Primoris is not solar at all. It is gas power generation. The company expects over $650 million of revenue there in 2026, up more than 30%, and Vadlamudi put a marker down on nine figures of growth: "over $1 billion in top line for next year." Backing it is roughly $900 million of verbal awards that could book in Q4 and a pipeline he pegged "in excess of $7 billion." Pipeline is turning too — a $250 million award burning at the end of 2027 into 2028, plus a $150 million job expected to contract and burn in 2027. The sales funnel hit a record above $8 billion. And then there is PayneCrest, the May acquisition that is now the quiet star. Backlog has gone from roughly $260 million at purchase to nearly $600 million; the company will clear $300 million of revenue in eight months, high end of the deal model. The exposure is industrial, including data centers, anchored on what Vadlamudi called “a very core hyperscale client that has an outlook for work that exceeds available capacity in terms of labor” — Koti Vadlamudi, President and Chief Executive Officer · 2026-11-04. That's the interesting inversion: hyperscale client demand is so strong the governor is crews, not contracts — the same constraint management cited in power delivery, where MSA work productivity drove gross profit to double the prior year and 13% segment margins in Q3. Here's the contrast worth holding onto: the market's AI-infrastructure complex is currently deflating hard — the global tape shows AI data center down more than 20% over 30 days with 56 decliners, and high bandwidth memory off roughly 30% over 90 days. Primoris is pitching data-center-adjacent growth into a tape that has stopped paying for it.Cash, leverage, and the promise that didn't hold
The weakest part of the story is cash. Ken Dodgen's admission was blunt: “we are now expecting operating cash flow of about $100 million for the year. And essentially breakeven free cash flow” — Kenneth Dodgen, Chief Financial Officer · 2026-11-04. Compare that with February's promise of “operating cash flow kind of in that 4% to 5% of revenue range... at least 50% of adjusted EBITDA, if not higher” — Ken Dodgen, Chief Financial Officer · 2026-02-24. On a ~$7 billion revenue base that's a gap of hundreds of millions. The trailing figure already shows it: free cash flow of -$34 million in the June quarter. Leverage explains the caution. PayneCrest was funded with debt, and effective net cash sits at -$577 million, versus a modest net cash position as recently as late 2025. Net debt to EBITDA is 1.8x and management expects it flat for a couple of quarters. That is why the $250 million buyback authorization is framed as "opportunistic" rather than a commitment, and why Vadlamudi's list of priorities starts with organic growth. One striking retreat: communications. After years of positioning fiber as a growth vector, including last November's talk of a “tremendous ramp up in the other types of fiber network build outs” — David King, Chairman and Interim President and Chief Executive Officer · 2025-11-04, Vadlamudi closed the door — BEAD is "not a meaningful growth vector for us," and fiber-to-the-home is "a diminishing market." That's a segment quietly written off.The one question everyone was circling
Reported the day after an election, the call kept returning to policy. Vadlamudi's answer on gas permits was the most useful thing said all morning:That matters because the entire 2027 recovery — backlog growth converting into profitable growth, energy margins back to 10%–12%, the modest growth framing in renewables, "exceed consensus" — depends on the contracted book staying contracted. Vadlamudi's own tell: if you strip out the six problem jobs, renewables margins are already back where they belong. The bull case is that 2027 is simply 2026 without the overruns. The bear case is that a 26x price-to-operating-income multiple on depressed earnings leaves no room for a second disappointment.Nothing we have currently has been delayed or canceled due to any moratorium. All the projects are in the field and moving forward with their current schedules.