OVHcloud: strong growth, thin cash flow and a 3% block sale, so hold at €13.68

OVHcloud finished its 2026 fiscal year with everything it promised: revenue of €1,145 million, growth that accelerated to 10.5% in the fourth quarter, an adjusted EBITDA margin of 42%, and a first positive levered free cash flow. The shares closed at €17.49 on results day, then fell 29% in three sessions after an unnamed shareholder sold 3% of the company at a 15% discount. The business is improving. The difficulty is that nearly all of its EBITDA is reinvested in servers, so very little reaches shareholders, and debt takes the first claim on the rest.

At €13.68 on October 29, 2026, the shares trade 25.4% above my Base value of €10.91, so the Base is 20.2% below the price. The Bear value of €4.27 is 68.8% below the price and the Bull value of €18.04 is 31.9% above it. I assign no probabilities to the cases. The required margin of safety of 50% gives a buy range of €3.61 to €7.30 and a conservative limit of €3.61. The verdict is Hold, and I would not add at this price.

Valuation route: cloud infrastructure provider with heavy server spending, valued on unlevered free cash flow after leases and capital spending, with enterprise value bridged to equity through net debt. Economic classification: quality growth with a capital-intensive model and below-cost-of-capital returns today. Primary method: ten years of free cash flow with a terminal value. Cross-check: 6.5 times estimated fiscal 2027 EBITDA. Values are present fair-value estimates, not 12-month price targets.

What OVHcloud is

OVHcloud, founded in France by Octave Klaba, who is again chairman and chief executive, rents servers, private cloud environments and web hosting from its own data centers. It has about 250 megawatts of data center capacity, of which about 140 megawatts is connected and about 110 megawatts is available. Management says it is now about two times cheaper than AWS in public cloud, down from three times.

Private Cloud is 60% of revenue, Public Cloud 22% and Web Cloud 17%. Public Cloud is the growth engine and Private Cloud is the large, slower base. The family of the founder is the controlling shareholder.

The latest year

For fiscal 2027 management guides to constant-currency growth above 10%, an adjusted EBITDA margin above 42.5%, positive levered free cash flow, return on capital employed above 9% and leverage of 2.7 times or lower. It removed its formal capital spending guidance and said spending will depend on free cash flow and returns. Price increases are included in the outlook and more are planned for November and December; management has not quantified them and wants first-quarter data on customer losses before it does.

What the cash flow is made of

• Almost every euro of EBITDA is spent. From €480.5 million of adjusted EBITDA, the company paid €422.4 million of capital spending, €39.3 million of lease repayments, €51.8 million of interest and €8.5 million of tax, and received €15.1 million from working capital.
• The headline free cash flow flatters. The €24.7 million adds back €49.5 million of fiscal 2027 purchases made early. That spending is real cash, and it lowers next year's need, but it does not make fiscal 2026 more profitable.
• Depreciation is close to capital spending. Operating profit is only 6.8% of revenue because depreciation takes roughly a third of it, and depreciation rose by €24 million in the year. The business does not have a large pool of old, fully depreciated servers that would let spending fall quickly.
• Adjusted EBITDA excludes real costs. The €20.0 million of share-based and earn-out costs is added back, and I deduct it in every value in this analysis.
• Margins will flow through less. Management expects only about 50% of extra revenue to become EBITDA in fiscal 2027, compared with 71% in fiscal 2026.

Unlevered free cash flow in fiscal 2026, after leases and a normal level of capital spending and tax, was about €28 million by my estimate, or roughly 6% of EBITDA.

Debt, shares and the recent block sale

I use 151.7 million shares and net debt of €1,157 million. Leases are treated as an operating cost, through the repayments I deduct from free cash flow, so I do not subtract the lease liability from enterprise value as well. The company reports equity-settled share plans costing €8.8 million; I did not model their dilution.

On the evening of October 22, an unnamed shareholder sold 4.65 million shares, about 3% of the capital, in an accelerated placement at €13.30, a discount of 15.3% to that day's close. The shares closed at €12.39 the next day. In December 2025 a vehicle controlled by the family sold 3.6 million shares, 2.3% of the capital, at €7.50. I could not find out who the October seller was or whether a lock-up applies, so another sale remains possible. A block sale does not change what the business is worth, but it shows that holders with large positions were willing to sell near these prices.

Debt is the lever that matters. Enterprise value is about €3.2 billion at the market price and debt is €1.2 billion of it, so a 10% change in enterprise value moves the share price by about 16%. Equity is thin as well: €74.5 million on the balance sheet.

Primary valuation: free cash flow

This takes revenue from €1,145 million to €1,616 million in fiscal 2030 and €2,104 million in fiscal 2036. EBITDA is the figure after share-based and earn-out costs; my 41.0% for fiscal 2027 corresponds to management's adjusted margin of more than 42.5%. Free cash flow is EBITDA less capital spending, lease repayments and cash tax at 25% of operating profit. Depreciation falls from 33.5% to 28.5% of revenue as growth slows, and I set terminal capital spending so that spending plus leases exceeds depreciation by about 2% of revenue, which is what 2.5% growth needs. I use a discount rate of 8.5%, a blend of a cost of equity of about 10.5% and a cost of debt of 4.4% after tax at the current balance sheet weights, and terminal growth of 2.5%.

• Result: an enterprise value of €2,251 million, less net debt of €1,157 million, gives equity of €1,094 million, or €7.21 a share on 151.7 million shares.
• Terminal value: 62.1% of the enterprise value, over a ten-year horizon. That is below the 75% attention level, but it sits on a terminal free cash flow margin of 8.7% that I cannot observe today.

No cell reaches €13.68. Other single changes to the Base also fall short: growth 3 percentage points higher every year gives €11.18, an EBITDA margin 2 points higher gives €9.95 and terminal growth of 3.5% gives €9.16. The price needs the terminal capital spending to fall to about 21% of revenue, which is the same as a terminal free cash flow margin of 14.8%.

Secondary valuation: EBITDA multiple

I estimate fiscal 2027 revenue at €1,265 million and EBITDA after share-based costs at €519 million. At 6.5 times this gives an enterprise value of €3,374 million, and after €1,157 million of net debt, €14.60 a share. At the current price the market pays 6.2 times the same figure, and 7.0 times fiscal 2026 EBITDA.

I chose 6.5 times because it is just below OVHcloud's own current multiple, to reflect that only a small part of EBITDA converts to free cash flow. This route therefore mostly tests whether the market's usual multiple still holds; it is not independent proof that the shares are cheap. I did not collect a dated peer set or a multi-year multiple history, so I make no claim about where the shares trade against other cloud providers.

The first method implies a multiple of only 4.3 times fiscal 2027 EBITDA, because in my Base just 20% of the terminal EBITDA becomes free cash flow.

Growth diagnostic

Revenue growth went from 5% in the second quarter to 11% in the fourth, and Public Cloud grew 24% in the fourth quarter. Monthly customer growth outside the United States rose from 0.7% early in the year to 2.0% by August. Price increases take effect in fiscal 2027, so they are not yet part of this acceleration. I classify the growth as structural in Public Cloud and moderate in Private Cloud, with the price component uncertain. The Corporates segment grew only 2.6% after the loss of two large customers in fiscal 2025.

The earnings are not yet following the growth. Net income of €9.4 million is 0.8% of revenue, and returns on capital of 7.7% are below my discount rate of 8.5%. The business adds value for shareholders only if returns exceed the cost of capital, and that point has not yet been reached.

Valuation triangulation

• Primary (free cash flow): €7.21, weight 50%.
• Secondary (6.5 times fiscal 2027 EBITDA): €14.60, weight 50%.
• Divergence between the two: 68% of their midpoint, above the 30% limit, so I flag a valuation divergence.
• Weighted Base: €10.91. The unweighted range of €7.21 to €14.60 is the honest uncertainty, and €10.91 is a point inside it.

The two methods share the same revenue and EBITDA forecasts, so their agreement would not have raised confidence, and their disagreement is about one thing only. The multiple value of €14.60 equals my cash flow model if terminal capital spending falls to about 20.5% of revenue (a terminal free cash flow margin of 15.6%) or if the discount rate is 6.6%. In other words, the market multiple assumes the company will need far less spending on servers than it does today. The weights are equal because I cannot show which view is right. Because the price sits between the two values, the choice of method changes the decision: the cash flow value says the shares are expensive and the multiple says they are fairly priced, so the range matters more than the single weighted figure.

What is the market already pricing in?

A buyer at €13.68 is paying for management's growth above 10% and for a business that, over time, needs much less capital spending per euro of revenue than it needs now. That is possible, since depreciation is already high and newer servers may be more efficient, but it has not been shown. There is no dividend to hold the price up while this is tested.

Bear, base and bull scenarios

• Bear: growth of 7% in fiscal 2027 fading to 2.5%, EBITDA margin of 40% rising only to 41%, capital spending of 33% falling only to 29% of revenue, for example because of higher hardware prices or customer losses after price increases, and 5.0 times fiscal 2027 EBITDA of €490 million. In the cash flow model the enterprise value of €1,122 million is below net debt, so the equity is worth about zero; I floor it at €0.00.
• Base: as described above.
• Bull: growth of 12% fading to 3.5%, EBITDA margin rising from 42% to 46%, capital spending falling from 31% to 25% of revenue, and 8.0 times fiscal 2027 EBITDA of €539 million.

The discount rate is the same in all three cases, because the scenarios describe the business and not the investor's requirement. The asymmetry is unfavorable: about 69% below the price in the Bear against 32% above in the Bull. Because of the debt, small changes in the business have large effects on the equity.

How the buy price is set

Confidence is D, which has a default margin of safety of 55% and a reference range of 50% to 65%. I use 50%, the low end, because Confidence is at the top of the D band and the divergence between methods is already part of the Confidence score and of the range. I add no further adjustment, so the margin is applied once, to each method.

The entry range is €3.61 to €7.30 and the conservative limit is €3.61. The price is well above both bounds, so there is no case for adding. This is not a call to sell: the multiple method puts fair value above the price, the Bull is 32% higher, and the Base is 20% lower, a spread that does not justify a forced exit by itself.

Dividend and buybacks

OVHcloud pays no dividend and describes no buyback programme in its fiscal 2026 release. Cash spent on treasury shares was €8.3 million in fiscal 2026 against €356.1 million in fiscal 2025. With free cash flow close to zero and leverage of 2.7 times, I do not expect shareholder distributions soon.

Reasons to own OVH

• All fiscal 2026 targets were met, growth accelerated in each half and management guides to more than 10% growth in fiscal 2027.
• Public Cloud, which grew 18.9% for the year and 24% in the fourth quarter, is becoming a larger part of the mix.
• Return on capital employed has risen from 3.3% in fiscal 2024 to 7.7%, and management targets above 9%.
• Price increases are under way, and management describes OVHcloud as still about two times cheaper than AWS in public cloud.
• The balance sheet has no major maturity before fiscal 2030 and a €200 million undrawn credit line.
• The company has about 110 megawatts of available capacity to sell into demand.

What could go wrong?

• Capital intensity does not fall. Capital spending was 32.6% of revenue and depreciation about a third, so profit and free cash flow stay thin even as revenue grows.
• Price increases may cause customers to leave. Management has not yet seen the data.
• GPU and AI demand is unproven. Management says it wants proof before committing large amounts, and capital spending guidance has been withdrawn.
• Component costs, especially memory, may rise, and management could not say whether it will buy fiscal 2028 components early.
• Debt of €1.2 billion amplifies any shortfall, and about €683 million falls due in fiscal 2030.
• The overhang from further share sales by large holders, as in October and December of last year.
• Political uncertainty in France was cited by management as a reason for some customers' hesitation.

Management execution

• The founder returned as chief executive and delivered the full set of fiscal 2026 targets, including constant-currency growth of 7.2% against a range of 5% to 7%.
• The company reached positive levered free cash flow and cut leverage, although the free cash flow figure relies on adding back early spending.
• Removing capital spending guidance lowers visibility for investors, and I treat it as a cost until spending is shown to fall.
• Three bolt-on acquisitions were made in AI software, with no amounts disclosed.

Stock Analyza scorecard

The valuation score is 26 of 100, in the overvalued band of 25 to 44.

Business quality and valuation are shown separately and are not averaged. A growing business of middling quality at a price somewhat above its central value is a hold and not an add.

Confidence score

The grade measures how well the model is identified, not the chance of a loss. The terminal capital spending, the debt leverage and the disagreement between methods are the reasons for the low grade.

What I would watch from here

The three most important thesis breakers

1. Capital spending stays near a third of revenue as growth slows, which would put free cash flow below my Base.
2. Price increases lead to lasting customer losses, which would remove the margin and growth guidance.
3. A fall in EBITDA of 10% or more, which at today's leverage would cut the equity value by far more than the business.

Stress test and independent check

I recomputed the values with a separate script and solved the reverse valuation for the discount rate and the terminal capital spending. That checks the arithmetic, not the forecasts. Several inputs are my estimates, as the table shows.

The weakest point of the model is the long-term capital intensity of the business. The strongest bull argument is that the servers bought in the last two years last longer than the depreciation assumes, that price increases lift margins without losing customers, and that capital spending falls to about a fifth of revenue, which would justify the current price. The strongest bear argument is that depreciation is already a third of revenue, the company has no cushion of fully written-down hardware, and its leverage and thin equity make any miss expensive for shareholders.

Final verdict: hold at €13.68

OVHcloud is a growing, improving business that is still earning less than its cost of capital and converts very little of its EBITDA into free cash flow. At €13.68 the price is between my cash flow value and my multiple value, and above the weighted Base.

Verdict: HOLD at €13.68, do not add. Weighted Base €10.91, 20.2% below the price, cross-check €14.60, Confidence D (12/25). Required margin of safety 50%, buy range €3.61 to €7.30, conservative buy limit €3.61.

For an existing holder, the original cost of the shares is irrelevant. I would hold a position of a size one can accept losing a large part of, because the leverage makes the downside wider than the business alone suggests, and I would review it when first-quarter fiscal 2027 results show whether growth, price increases and capital spending are moving as management says. New money waits for a price at or below €7.30, or for evidence that capital spending is falling.

Sources

Primary sources are OVHcloud's fiscal 2026 results release of October 20, 2026, the earnings call transcript and a press summary of the results presentation, which I used for segment, customer and maturity detail and which the publisher notes was prepared with automated assistance. The share count is from the company's June 30, 2026 voting rights statement. The October 2026 block sale is from ABC Bourse and Boursorama, and the December 2025 sale from EasyBourse.

The reference market price is €13.68, the October 29, 2026 regular session close, from MarketScreener. A broker screenshot on October 30 shows €13.22, which is an intraday figure 3.4% lower and is not a regular session close. Balance sheet figures are as of August 31, 2026. Closing prices around the results are from ABC Bourse.