GE Vernova is an exercise in separating the economics of a complicated industrial company. Growing electricity demand may be helpful, but the shareholder outcome depends on which equipment customers order, what prices and contractual protections apply, and how successfully the company delivers. A broad electricity thesis is insufficient.
The company became independently traded on April 2, 2024. Its Power, Wind and Electrification businesses therefore now have a distinct public valuation and capital allocation framework. Separation creates an opportunity to study those businesses directly, but it does not itself make the shares cheap. Spin-off completion.
Watch it play outOne spin-off, three businesses to value
GE Vernova's businesses used to sit inside GE, valued as part of a bigger whole.
On April 2, 2024, GE Vernova became independently traded, with its own share price and capital allocation.
Open it up and there are three businesses: Power, Wind and Electrification.
Separation lets investors study each one directly. It does not, by itself, make the shares cheap.
The January 28 release reports 2025 revenue of $38.1 billion, up 9%, adjusted EBITDA margin of 8.4% and free cash flow of $3.7 billion. GAAP net income included a $2.9 billion tax benefit. Gas Power backlog plus slot reservations reached 83 GW. Management's 2026 outlook is $44–45 billion of revenue, 11–13% adjusted EBITDA margin and $5.0–5.5 billion of free cash flow, including the expected Prolec GE transaction. Guidance is a forecast; the acquisition is expected to close February 2 and is not yet completed at this date. January 28 results.
Watch it play outA big business on thin margins, guided to widen
GE Vernova's 2025 revenue was $38.1 billion. Watch it grow from the year before.
That was 9% growth. The slice kept as adjusted EBITDA is small: an 8.4% margin.
Management guides 2026 to $44–45 billion of revenue at an 11–13% margin. Guidance is a forecast.
Source: GE Vernova results, January 28, 2026
What the market may be missing
There are two possible valuation mistakes. One is to treat the entire company as a low-margin legacy equipment supplier. The other is to assign every segment the economics of its strongest business. A good thesis must resist both.
Industrial companies often recognize the consequences of earlier commercial decisions over many years. An order booked at a poor price can remain a burden even when current demand is strong. New orders may be better, but their contribution arrives gradually. An investor who looks only at present margins can miss improvement; an investor who ignores the old backlog can overstate its speed.
Two easy valuations, both wrong
Low-margin legacy supplier
Treats the whole company as legacy equipment
Misses improvement in new orders
Best segment everywhere
Gives every segment its strongest economics
Overstates the speed of improvement
Segment by segment
Age and quality of the backlog
New orders priced for delivery risk, inflation and guarantees
This makes the age and quality of the backlog central. The relevant question is not simply whether orders exceed revenue. It is whether new business earns enough to compensate for delivery risk, inflation, guarantees and the capital required to fulfill it. Larger backlog can increase both expected profit and the amount that can go wrong.
The first- and second-quarter 2024 releases provide earlier financial checkpoints. That second-quarter announcement raised 2024 guidance, so improving expectations are already public information. This note does not assume the company is undiscovered, and it does not claim to know historical analyst consensus. First-quarter results, second-quarter results.
The underlying industry shift
Electricity investment has several different layers. Generation produces power. Transmission moves it. Distribution connects customers. Equipment at the customer site then converts it into useful work. A bottleneck at any layer can prevent demand at the final layer from translating into operating consumption.
Demand has to clear every layer of the grid
GenerationProduces power
TransmissionMoves it
DistributionConnects customers
Customer-site equipmentTurns it into useful work
A bottleneck anywhere stops demand
For a manufacturer, the opportunity comes from selling equipment that solves a funded customer problem. A data center owner's ambitious plan has limited immediate relevance if the owner lacks interconnection, permits or a credible supply arrangement. Likewise, an electricity forecast does not specify which generating technology will receive orders or when equipment will ship.
Gas equipment offers one possible route to dependable generation. The economic attraction for a supplier could extend beyond the initial sale if installed equipment creates recurring service requirements. That recurring component should be evaluated separately from the equipment margin. An order may create a long customer relationship, but the value of that relationship depends on contract terms, utilization and service delivery costs.
Grid equipment presents another route. If customers need electrical infrastructure regardless of their preferred generation technology, exposure to the connecting network can diversify a power-equipment thesis. That is a structural inference, not a prediction that every grid project will be approved or profitable.
The strongest investment case would combine favorable new equipment economics with a growing service base and improving execution. The weakest would rely on a rising industry budget while ignoring the terms on which the manufacturer accepts the work.
Why GE Vernova could benefit
The proposed advantage is an existing industrial platform that can participate across several stages of power investment. The challenge is that the same breadth makes headline comparisons misleading. A strong order trend in Power does not automatically repair Wind, and cash arriving on one contract cannot be interpreted as profit on another.
A practical model should therefore keep three ledgers. The first tracks new orders and their likely delivery period. The second tracks installed equipment and service activity. The third tracks obligations embedded in older contracts. Collapsing these into a single revenue growth assumption removes the very information that might create an analytical advantage.
Consider an illustrative equipment contract. Receiving a customer deposit improves current cash flow, but the supplier still owes manufacturing, logistics and commissioning. If the final cost rises, that earlier cash inflow does not protect the economic margin. Conversely, a temporary cash outflow during production need not signal weak profitability if the contract remains appropriately priced and collectible.
This distinction changes how one reads free cash flow. A strong quarter is useful evidence of liquidity. It is not a license to capitalize the quarter four times and call the result normal annual earnings. Working capital needs to be evaluated across a delivery cycle.
The numbers
The archived Yahoo Finance price series records $717.39 at the January 29, 2026 close. Multiplying by 270 million Q4 weighted-average basic shares gives approximately $193.70 billion of equity value. This uses a lagged average count, not an exact contemporaneous capitalization. Full-year diluted EPS was $17.69, so trailing P/E is approximately 40.6; the tax benefit makes that multiple a poor measure of recurring earnings. Historical prices, January 28 financial tables.
Full-year gross profit of $7.535 billion on $38.068 billion revenue implies a 19.8% gross margin. Operating income of $1.388 billion implies 3.65% operating margin. Those GAAP measures differ materially from 8.4% adjusted EBITDA margin. Reported $3.710 billion free cash flow is approximately a 1.92% cash yield on the lagged-count equity proxy. A historical consensus forward P/E and acquisition-adjusted EV/EBITDA remain unavailable; a pending $5.275 billion acquisition makes a casual net-cash subtraction especially misleading. Financial statements and transaction update.
Three margins tell three different stories
GAAP gross margin19.8%
Adjusted EBITDA margin8.4%
GAAP operating margin3.65%
Full-year 2025. Trailing P/E of about 40.6 rests on EPS lifted by a $2.9 billion tax benefit.Source: GE Vernova financial statements, January 28, 2026
The available annual revenue and cash-flow figures are enough to demonstrate why simple trailing P/E is a poor standalone lens. A tax benefit can inflate annual GAAP earnings without improving recurring economics, and positive free cash flow can partly reflect timing. Neither observation proves undervaluation.
A proper enterprise-value model must also distinguish unrestricted cash, debt, contractual obligations and other claims. Subtracting every dollar of reported cash from market capitalization would overstate financial flexibility if some of that liquidity is needed to execute prepaid contracts. The appropriate reserve is a modeling judgment that should be made visible.
The missing historical valuation inputs limit the conviction of this case study. The company can merit detailed work without meeting the standard for a verified asymmetric entry price.
Valuation and the burden of proof
Management's 2026 revenue and margin outlook implies approximately $4.84 billion to $5.85 billion of adjusted EBITDA by multiplying the low and high endpoints. That arithmetic is derived from guidance, not a guarantee or a consensus forecast. It also includes the expected acquisition, so comparisons with the old perimeter require care.
For an independent stress test, assume $44 billion of revenue and an 8% EBITDA margin, below the guided range. That produces $3.52 billion of EBITDA. A stronger assumed 13% margin on $45 billion produces $5.85 billion. The large spread shows how sensitive the thesis is to conversion of revenue into profit.
Margin conversion decides GE Vernova's 2026 EBITDA
Revenue × margin. $44 billion to $45 billion at 11% to 13% is management's 2026 outlook, including the expected Prolec GE deal; 8% is the analyst stress case. At 20 times, $3.52 billion and $5.85 billion imply $70.4 billion and $117.0 billion of enterprise value.
At an assumed 20 times enterprise-value multiple, those cases imply $70.4 billion and $117.0 billion of enterprise value before the equity bridge. This is deliberately not a price target: acquisition financing and the exact point-in-time share count have not been fully rebuilt. It exposes the value placed on durable margin improvement.
A reverse model starts with verified enterprise value and asks what margin is required at a conservative multiple. If the answer demands immediate best-case execution across all segments, protection from disappointment is limited. Better orders may take years to dominate revenue, and delayed improvement leaves more time for inflation or execution problems to interfere.
Catalysts
Future earnings releases can provide evidence about the mix of orders, execution of existing projects and progression of service profitability. An increase in orders is most valuable when management also provides enough detail to assess conversion and commercial quality.
Changes in longer-term targets can be informative, but targets are promises rather than completed economics. The better test is whether intermediate operating milestones support them. Examples include delivery reliability, improving segment margins and fewer adverse contract adjustments. Shareholders should reward improving evidence, not merely more ambitious language.
Capital allocation also matters after separation. The company can retain cash for operating resilience, invest in productive capacity or return excess capital. The right choice depends on the return available from reinvestment and the obligations already assumed. A repurchase at an excessive valuation can destroy value even while increasing EPS.
Customer announcements should enter the model only after contractual status is understood. A memorandum, reservation, firm order and completed delivery are different points in the economic chain. Assigning the same probability and margin to all four would manufacture certainty.
Four kinds of customer announcement, four probabilities
Memorandum to Reservation
Reservation to Firm order
Firm order to Completed delivery
Each step should carry its own probability and margin in the model.
What could go wrong
The central risk is that rising demand creates activity without acceptable returns. Equipment can be essential while the supplier earns disappointing margins because contracts are competitive, execution is difficult or customers have bargaining power. Industry importance is not an economic moat by itself.
A second risk is that weak legacy contracts consume more cash than the stronger businesses produce. Diversification is helpful only if the losses are bounded. Investors should resist treating every charge as exceptional when the business repeatedly encounters new problems.
A third risk is overexpansion. If suppliers add capacity based on optimistic forecasts and customers later defer projects, the industry can move from scarcity to aggressive pricing. Production assets can then lower returns rather than strengthen the franchise.
A fourth risk is financial interpretation. A large cash balance and positive quarterly free cash flow can obscure future delivery obligations. Likewise, adjusted profit measures can clarify underlying operations or make recurring problems appear smaller, depending on what is excluded. The reconciliation deserves the same attention as the headline.
The thesis should weaken if order quality deteriorates, if improved Power profitability fails to reach consolidated results, or if working-capital benefits reverse without corresponding delivery progress. A lower stock price alone does not settle those questions; operating evidence does.
The investment decision
GE Vernova deserves analysis as a portfolio of contractual economics. The plausible opportunity is that profitable power and grid activity can outweigh legacy burdens as new orders become revenue and equipment becomes service demand. The case is strongest when conservative assumptions support an acceptable return.
Make a guess
Spot on.Close, a little low.Close, a little high.Way off, too low.Way off, too high. It was +31.3%.
GE Vernova closed at $941.95 on September 4, 2026, about 31.3% higher. Most of its earlier rally had already happened, which is why the look-back calls this a continuation case.
GE Vernova+31.3%
SolarEdge, same dates+0.5%
Enphase, same dates−4.9%
Source: Historical closing prices
This note establishes the mechanism and the questions to ask. It does not establish a fully verified bargain because essential historical valuation inputs remain incomplete. The next step is a segment model with an explicit cash-obligation bridge, not a higher price target based on a larger electricity forecast.
Looking back: what happened next
Historical information available: Separation was completed; full-year 2025 results and 2026 guidance were public; the expected Prolec GE close remained in the future.
Still unknown: The actual Prolec GE closing, later gas-equipment reservations, new contracts, guidance changes, segment results and stock prices.
The actual bet: Better commercial terms and a growing service base would turn electricity investment into higher consolidated cash earnings despite legacy contract burdens.
What actually happened: Q2 2026 revenue reached $11.1 billion, adjusted EBITDA margin 11.3%, and gas backlog plus reservations 116 GW. That supports continued demand and conversion, although customer advances complicate interpretation of exceptionally strong quarterly cash flow. July 22 release.
Return measurement: September 4, 2025 close $598.81 to September 4, 2026 close $941.95 gives 57.3% price appreciation. The January 29 close of $717.39 to that endpoint gives 31.3%. These are price returns excluding dividends and costs. Yahoo price history, independent endpoint.
GE Vernova had already run before the call
Sep 4, 2025$598.81
Jan 29, 2026 analysis$717.39
Sep 4, 2026$941.95+31.3%
$600$800$1,000
Price only. September 4, 2025 to September 4, 2026: +57.3%.Source: Yahoo price history
Timing: late, not early. The company's earlier major rally began well before January 2026. This analysis date is approximately seven months before the endpoint, but that is not seven months before the rally began. It is a continuation case with a positive subsequent return, not a first discovery.
Survivorship controls: Between January 29 and September 4, 2026, Enphase fell from $38.25 to $36.37, approximately 4.9%, while SolarEdge moved from $34.04 to $34.20, approximately 0.5%. Both lagged GEV's 31.3% gain. These are broad electrical-equipment controls with different customer and technology exposure, not gas-turbine substitutes. Their results show why a generic electricity theme is inadequate, but do not isolate the effect of service backlog or contract terms. Enphase history, SolarEdge history. Vestas and Siemens Energy are closer proposed equipment controls; their matched returns remain unverified.
A generic electricity label did not produce GE Vernova's return
GE Vernova+31.3%
SolarEdge+0.5%
Enphase−4.9%
Price returns, January 29 to September 4, 2026. Broad electrical-equipment controls, not gas-turbine substitutes.Source: Historical closing prices
NVIDIA has found a way to turn knowledge of its own roadmap into a lower cost of capital for its customers. The result could move AI infrastructure from venture equity into private credit, insurance balance sheets and securitized debt.
The thesis, the numbers behind it, and what would break it. Full access is $39 a month.
