IRM — Iron Mountain Incorporated

The analyst community is still constructive - 14 Buy vs. 2 Hold and 5 Sell, with a consensus price target of $146.33 (high $155, low $140, n=4 for the recent targets) - and the Aug 5, 2026 quarter was a genuine beat-and-raise (revenue $2.03B, +18.5% YoY, vs. ~$1.97B consensus; adjusted EPS $0.60; AFFO $1.44/share). But the tape disagrees with the sell side right now: as of 2026-09-13 IRM was -7.38% over one month and -9.48% over three months, RSI was 42.06, and early-September Form 4/144 filings showed planned insider sales. The market is no longer paying up for the growth story - it is re-pricing the funding of that story. Sentiment therefore reads as Neutral: bullish fundamentals, cautious price action.

Revenue growth is clearly ACCELERATING, not decelerating: YoY growth ran +7.8% (Q1'25), +11.6% (Q2'25), +12.6% (Q3'25), +16.6% (Q4'25), +21.6% (Q1'26), +18.5% (Q2'26) - revenue went from $1.39B in Q3 2023 to $2.03B in Q2 2026. Operating margin is expanding in step: 15.2% in Q2'25 vs. 18.4% in Q2'26 (+320bps), and 16.0% vs. 20.4% in Q1 (+440bps). Gross margin is moving the other way - 55.9% in Q2'25 vs. 52.8% in Q2'26 - which tells you the mix is shifting toward lower-gross-margin Asset Lifecycle Management (~$1B run-rate, +58% YoY) and data centers, with operating leverage below the gross line absorbing the difference. GAAP earnings remain lumpy and low quality-looking: net income was negative in Q3 2024 (-$33.6M) and Q2 2025 (-$44.9M), and TTM net margin is only 5.5%. The structural feature that matters most is cash: free cash flow was negative in 11 of the last 12 quarters, cumulating to roughly -$1.88B, while capex ran about $5.97B against $4.09B of operating cash flow over the same window. That is not deterioration - it is a deliberate build-out financed with debt ($11.8B long-term debt at 2023-09-30 to $17.13B at 2026-06-30) - but it does mean every quarter of growth is currently being bought with borrowed money.

As of 2026-09-13, IRM was not cheap. The headline P/E of 82.3x (on TTM GAAP EPS of $1.41) is close to useless for a REIT carrying heavy depreciation, and the 0.091 PEG is distorted beyond use. More honest: forward P/E of ~46x on next-FY EPS of $2.494, P/S of 4.53x vs. a 5-year average of 3.70x (a ~22% premium to its own history), EV/EBITDA of ~21.5x, and a 2.67% dividend yield that is thin for a REIT and signals reinvestment over income. Price-to-book is negative (-26.8x) because stockholders' equity is -$1.28B, so book value gives no floor. My own work (see theoretical fair price) puts fair value near $123, so the shares look roughly fairly valued with low-double-digit upside rather than either a bargain or a bubble - the multiple is full and it is leaning hard on continued mid-teens growth.

As of 2026-09-13, the price is falling while estimates are rising - which is the opposite of the usual 'pricing in growth that hasn't shown up yet' problem. Guidance was raised on 2026-08-05, and the stock is still -9.48% over three months and -7.38% over one month. That means the de-rating is not about earnings at all; it is about how the earnings are financed. The market is applying a higher discount rate to capex-heavy, debt-funded growth. Practically: if the Nov 4 Q3 report confirms data center leasing and ALM momentum, this compressed multiple is the opportunity, because you are buying improving numbers at a lower price than three months ago. If instead the story stays on leverage and negative free cash flow, expect multiple compression to keep doing the work regardless of how good the revenue line looks. Watch the balance sheet and capex line as closely as the top line.

As of 2026-09-13, IRM was in a short-term corrective downtrend inside a longer-term uptrend. RSI(14) at 42.06 is neutral-to-weak - not oversold, so there is no capitulation signal to buy. Price of $115.18 was 3.59% below the 20-day SMA (~$119.47), 5.14% below the 50-day SMA (~$121.42), but still 4.58% above the 200-day SMA (~$110.14), within a 52-week range of $77.77-$134.68. The 200-day near $110 is the level that matters: holding it keeps the multi-year uptrend intact and would make a test of that zone with RSI in the low 30s a reasonable entry; a decisive break below it would confirm the correction has more to run. Timing implication: no urgency to chase at $115 with RSI at 42 - the setup is 'wait for the 200-day test' rather than 'buy the dip.'

• Interest rates and refinancing cost: IRM carried $17.13B of long-term debt against -$1.28B of stockholders' equity as of 2026-06-30, up from $11.8B at 2023-09-30. Every 100bp on the curve directly hits the cost of funding the data center build, and the dividend yield of 2.67% offers little cushion versus a rising risk-free rate.
• AI and hyperscaler capex cycle: over 70% of active data center expansions are already pre-leased, which ties IRM's fastest-growing, highest-multiple segment directly to hyperscaler and AI compute budgets. A capex digestion phase would hit the best part of the story first.
• REIT sector rotation and the cost of capital: after a 36-48% YTD run per the research brief, money has been rotating out of high-multiple infrastructure REITs. IRM's modest 2.67% yield makes it a growth REIT, not an income REIT, so it is exposed to that rotation.
• AI hardware refresh and decommissioning cycles: Asset Lifecycle Management is tracking toward roughly $1B of 2026 revenue (+58% YoY) as enterprises retire servers and AI accelerators - durable, but demand is tied to corporate IT capex rather than IRM's own pricing power.
• Power, land and construction constraints on data center capacity: grid interconnection and supply-chain delays can push capex out and delay the revenue it is meant to produce, which matters because capex ($5.97B over 12 quarters) is running well ahead of operating cash flow ($4.09B).

• Q3 2026 earnings, expected Wednesday, November 4, 2026 pre-market - the key read on data center leasing, ALM scaling, margin trajectory, and whether FY26 revenue guidance of $7.94-8.01B gets raised again.
• Data center leasing velocity: with over 70% of active expansions pre-leased, any new hyperscale signings or capacity announcements support the multiple even before revenue lands.
• Asset recycling or a data center joint venture: an announced JV or minority-stake sale would directly attack the negative free cash flow and leverage concern that is the core of the bear case.
• An easier rate path: a Fed easing cycle lowers IRM's cost of capital and widens the spread between AFFO and cap rates, mechanically lifting REIT valuations.
• International expansion of the InSight DXP AI document-processing platform, including the September 2026 partnership with Saudi Arabia's Social Development Bank - small today, but evidence the digital services layer can add higher-margin revenue on top of storage.

• Balance sheet leverage: $17.13B of long-term debt, -$1.28B of stockholders' equity, a 0.726 current ratio, and reported ROE of -14.7%. Long-term debt has grown about 45% since 2023-09-30 and is still climbing.
• Negative free cash flow: FCF was negative in 11 of the last 12 quarters, cumulating to roughly -$1.88B, with capex of about $5.97B against $4.09B of operating cash flow. Growth depends on continued access to debt or equity markets at acceptable rates.
• Earnings quality and volatility: GAAP net income was negative in Q3 2024 (-$33.6M) and Q2 2025 (-$44.9M), TTM net margin is just 5.5%, and ROIC is 5.3% - low for the cost of capital being taken on.
• Gross margin mix compression: gross margin fell from 55.9% (Q2'25) to 52.8% (Q2'26) as lower-margin ALM and data center revenue outgrew storage; operating leverage has absorbed it so far, but that is an assumption, not a guarantee.
• Full valuation with no valuation cushion: at $115.18 on 2026-09-13, forward P/E was ~46x and P/S was 4.53x versus a five-year average of 3.70x. If growth decelerates even modestly, the multiple has room to compress much faster than earnings can grow.

Iron Mountain is a genuinely accelerating business - revenue growth has gone from +7.8% to +18.5% YoY over six quarters, operating margin is expanding by 300bps+, guidance was raised on Aug 5, and over 70% of data center capacity under construction is pre-leased - wrapped around a balance sheet that is the opposite of comfortable: -$1.28B of equity, $17.13B of debt, and negative free cash flow in 11 of the last 12 quarters. At $115.18 on 2026-09-13 the shares are roughly fairly valued by my math (~$123), not a bargain, and the technical picture is a stock in a corrective pullback with no oversold signal. This is a legitimate long-term infrastructure story with a real financing risk attached, and the current price reflects most of the good news - own it for the data center and ALM ramp, size it for the leverage.

I used an EV/EBITDA approach, the right tool for a leveraged, capital-intensive REIT. TTM EBITDA implied by the data is ~$2.38B (a ~$51.2B EV at the quoted 21.5x, consisting of $34.27B market cap + $17.13B long-term debt - $0.205B cash); growing that at FY26 revenue guidance of $7.94-8.01B (midpoint $7.975B) at a ~32% margin gives ~$2.55B of FY26E EBITDA. Applying 21x - a slight discount to the current 21.5x to reflect negative equity and negative free cash flow, but a premium to legacy storage REITs given the pre-leased data center pipeline - yields an EV of $53.6B, less $16.92B of net debt, for $36.6B of equity, or ~$123/share on 297.5M shares. Cross-check: ~22x FY26E AFFO of roughly $5.60/share (Q2 2026 AFFO was $1.44/share) also lands near $123.