# Globant: cheap on adjusted earnings, expensive on the cash that reaches owners GLOB, NYSE. Published October 21, 2026. Price as of the October 20, 2026 regular session close. Values in US dollars. ## Verdict: HOLD at $41.65 At $41.65 the shares trade 13.4% above my Base value of $36.73, a downside of 11.8%. Globant looks cheap at 7.0 times its guided adjusted earnings, but after investment, stock compensation, leases, interest and minority holders the cash left for shareholders is about $125 million a year in my Base, against adjusted profit of about $259 million. Confidence is C and the required margin of safety is 45%, which puts the maximum buy price at $20.20. Hold for holders, watchlist for new money. - Bear value: $16.20 - Base value: $36.73 - Bull value: $58.49 - Downside to base value: -11.8% - Maximum buy price: $20.20 - Required margin of safety: 45% - Price to guided adjusted earnings: 7.0 times - Owner cash, starting level: $125 million a year - Modeled annual return: 4.7% - Business quality: 51/100 - Valuation score: 38/100 - Confidence: 13/25, grade C Globant trades at 7.0 times the midpoint of its adjusted earnings guidance, which is the number a screener shows you. The cash left for shareholders in 2025, after stock-based compensation, lease payments and interest, was $79.8 million on $2,454.9 million of revenue (my reconstruction from the audited cash flow statement). Those two facts describe the same company. At $41.65, the October 20, 2026 regular session close, the shares trade 13.4% above my triangulated Base value of **$36.73**, a downside of 11.8%. The Bear value is $16.20, 61.1% below the price, and the Bull value of $58.49 is 40.4% above it. Confidence is grade C (13/25), the required margin of safety (MOS) is 45%, and the maximum buy price is **$20.20**. The price is above my Base and about twice the buy ceiling, so the verdict is Hold. Valuation route: an IT services and digital engineering group that sells engineers' time and outcome-based projects, valued on the cash that reaches parent shareholders after all investment, stock compensation, leases, interest and minority holders. Primary method (65% weight): three years of that owner cash, discounted at 10%, plus an exit value of 11 times the third year. Secondary method (35%): 10 times sustainable operating profit after stock compensation and acquired intangible costs, converted to equity with the same debt and claims. Classification: standard company, 4 of 12 on the scorecard. ## What Globant is Globant S.A. is a Luxembourg holding company that reports in US dollars under IFRS and files a Form 20-F. It sells software engineering, data, design and integration work, and in the second quarter of 2026 North America was 52.8% of revenue. Revenue in 2025 was $2,454.9 million, up 1.6% from $2,415.7 million. IFRS operating profit was $171.7 million, a 7.0% margin, against an adjusted margin of 15.2%. The gap is made of items Globant adds back, mostly stock compensation, restructuring charges and amortization of acquired intangibles. Growth has stopped. Second quarter revenue was $614.4 million, flat against $614.2 million a year earlier, and first half revenue was $1,221.5 million against $1,225.3 million. Customers above $1 million of trailing twelve month revenue fell from 339 to 331, and the ten largest customers rose from 29.3% to 30.6% of revenue. These are signs of a narrower customer base, not of distress. The new story is AI. Globant's Glob.AI platform had $52.8 million of annualized recurring revenue in the second quarter, up 61% on the quarter, and the company expects at least $110 million by the end of 2026. Against guided revenue of about $2,445 million, $52.8 million is 2.2%. An annualized run rate is not recognized revenue, and I do not add a software multiple to a services company because one of its products is growing fast. The open question is whether AI lets Globant deliver more work with fewer engineers and keep the saving, or whether clients take the saving in lower prices. ## The latest results | Latest reported period | Result | | --- | --- | | Second quarter 2026 revenue | $614.4 million, flat year over year | | Second quarter IFRS operating profit | $19.9 million, a 3.2% margin | | Second quarter adjusted operating margin | 13.2% ($80.9 million); first half 13.6% | | Second quarter business optimization charge | $32.3 million, against $47.6 million a year earlier | | Second quarter operating cash flow | $30.2 million; Globant free cash flow $12.6 million, against -$2.9 million a year earlier | | First half free cash flow (operating cash flow less capital expenditure) | $48.7 million, against -$8.6 million | | Second quarter diluted weighted shares | 43.228 million, down 2.4% from 44.298 million | | Cash and short-term investments, June 30 | $168.8 million | | Borrowings, June 30 | $422.0 million ($19.4 million current) | | FY2026 guidance, August | Revenue $2,428 to $2,462 million; adjusted EPS $5.75 to $6.15 | | FY2026 guidance, February and May | Revenue $2,460 to $2,510 million; adjusted EPS $6.10 to $6.50 | The August guidance cut the midpoint of revenue by 1.6% and the midpoint of adjusted EPS by 5.6% (my arithmetic). Globant's own free cash flow improved sharply, which is real, but as the next section shows it leaves out costs that a shareholder still pays. ## Which cash to value Globant's cash flow statement puts interest on borrowings and lease payments in financing, so operating cash flow minus capital expenditure leaves both unpaid. Capitalized development is a real investment even though it is booked as an intangible asset. And the statement adds stock compensation back, which is only fair if the shares issued cost shareholders nothing. I value the cash that reaches parent shareholders after all of these, and I count stock compensation as a cash cost at its recent level. | $ million, reconstructed from the cash flow statements | FY2023 | FY2024 | FY2025 | | --- | --- | --- | --- | | Operating cash flow | 318.5 | 248.7 | 301.2 | | Cash property and intangible investment | -126.8 | -113.6 | -89.5 | | Stock compensation added back by the statement | -72.9 | -81.8 | -75.7 | | Lease principal | -38.5 | -36.6 | -28.6 | | Lease interest | -6.3 | -6.9 | -6.8 | | Interest on borrowings | -4.1 | -12.2 | -20.8 | | Cash left before minority holders | 69.8 | -2.5 | 79.8 | In 2025 the cash left was 37.7% of operating cash flow less investment ($211.7 million). The three-year average is $49.1 million. I show that average on purpose, because a recovery thesis can quietly pick the best year. The strong 2025 year-end collections should not be multiplied by four, and 2024 was held down by working capital and investment. For the latest twelve months I roll the same bridge forward with the first half of 2026. Interim capital expenditure is a cash versus accrual mix, so every figure in this table is my estimate and not a reported number. | $ million, twelve months to June 2026 (my estimate) | Amount | | --- | --- | | Operating cash flow less investment, rolled forward | 269.0 | | Stock compensation as a cash cost | -71.7 | | Lease principal and interest | -35.4 | | Interest on borrowings | -27.0 | | Minority holders' economic share | -7.0 | | Cash for parent shareholders | 127.9 | I start the forecast at **$125 million** a year, a rounded figure with a range of $100 million to $150 million. It is well above the historical average, so it already gives Globant credit for the recovery in operating cash flow. The $7 million for minority holders is a modeled share of cash, not a known dividend, because accounting profit of minority holders is not cash paid out. ## Cash flow before and after financing Debt matters here. Borrowings of $422.0 million include maturities on May 30, 2028, inside my three-year horizon, and the annual report sets covenants of at least 3 times interest cover and at most 3.5 times net leverage on contractual definitions. Interest is about 17% of the cash available before financing, so deducting interest alone would not prove the equity value. The table shows the cash at each level, and what happens to debt. | Base case, $ million | Year 1 | Year 2 | Year 3 | | --- | --- | --- | --- | | Operating cash flow after tax and working capital | 345 | 369 | 394 | | Cash investment | -79 | -82 | -85 | | Stock compensation as a cash cost | -72 | -75 | -78 | | Lease principal and interest | -35 | -35 | -35 | | **A: before financing** | **159** | **177** | **196** | | Interest on borrowings | -27 | -29 | -32 | | **B: after interest** | **132** | **148** | **164** | | Minority holders' share | -7 | -8 | -9 | | **C: owner cash after net borrowing** | **125** | **140** | **155** | | Net borrowing or repayment | 0 | 0 | 0 | | Borrowings at year end | 422.0 | 422.0 | 422.0 | | Cash and short-term investments | 168.8 | 168.8 | 168.8 | | Net debt | 253.1 | 253.1 | 253.1 | | Net debt divided by level A | 1.6 times | 1.4 times | 1.3 times | | Interest as a share of level A | 17.0% | 16.4% | 16.3% | The refinancing assumption is explicit: the whole $422.0 million is replaced when it matures, so debt stays flat while the interest bill rises from $27 million to $32 million, or 6.4% to 7.6% of the balance (my assumption). Owner cash is treated as available to shareholders and not added a second time, so net debt stays at $253.1 million. If the cash accumulated instead, net debt would fall, but I would not add it to the value. Acquisition claims are paid outside owner cash and deducted once in the bridge below. A 300 basis point higher rate on the refinanced debt costs about $12.7 million a year before tax, or $9.5 million after an assumed 25% tax shield. From year 2 that lowers the Base by $2.14 on the cash method and by $1.39 in the weighted value, to $35.34 (my arithmetic). The ratio to cash flow before financing is not the covenant ratio. Globant does not report EBITDA and the inputs I have do not include depreciation, so I cannot test the covenants, and October headroom is unknown to me. A failure to refinance is a separate case and I have not modeled it. ## Shares, claims and the equity bridge I use 43.6 million diluted shares, the average Globant guides for 2026. The second quarter diluted weighted count of 43.228 million is 2.4% below a year earlier, and a data page shows about 43.2 million shares outstanding. I do not have an October diluted count. At $41.65 the model market value is $1,815.9 million. Stock compensation is charged as cash, so I hold the share count flat and do not also dilute it. | Equity bridge, $ million | Amount | | --- | --- | | Borrowings, June 30 | 421.955 | | Acquisition claims excluding minority holders (my estimate, range 80 to 148) | 120.000 | | Minority holders at book value | 60.375 | | Less excess cash (168.815 less 100 operating reserve) | -68.815 | | Net claims between enterprise value and common equity | 533.515 | At June 30 Globant reported $148.0 million of other financial liabilities, $97.8 million current and $50.2 million non-current, plus $95.9 million of lease liabilities. I treat $120 million as what is owed to the sellers of acquired businesses, apart from options over minority stakes, and $60.4 million as the book value of those minority stakes. A purchase option over a minority stake is another way of settling the same interest, so I do not add it again. Nobody outside Globant can split the $148.0 million with certainty, and the $100 million operating cash reserve is my judgment. Leases are charged in cash in the cash method, so I do not deduct the lease liability again. Goodwill makes book value a weak anchor. Parent equity after goodwill and intangibles is about $206 million, roughly $4.72 a share, from the supplied analysis. That is not a liquidation value. The new $125 million buyback authorization permits purchases and is not a completed purchase. In the second quarter Globant spent $25 million and finished the old program. I take no credit for future buybacks, because using the same cash for repurchases and for owner cash would count it twice. ## The primary valuation: three years of owner cash and an exit value The primary method discounts the three years of owner cash at 10% (a time-value assumption, tested at 8% and 12%) and adds a value at the end of year three equal to 11 times that year's cash. The exit value is a future figure, not today's value, so it is discounted too. | Base case | Owner cash, $ million | Present value, $ million | | --- | --- | --- | | Year 1 | 125.0 | 113.6 | | Year 2 | 140.0 | 115.7 | | Year 3 | 155.0 | 116.4 | | Exit value, 155 times 11 | 1,705.0 | 1,281.0 | | Plus excess cash, minus acquisition claims of 120 | | -51.2 | | Equity value today | | 1,575.6 | | Per share, 43.6 million | | $36.14 | The primary Base is **$36.14**. The exit value is 78.7% of the value of the operating cash flows, or 81.3% of equity value after excess cash and claims. The explicit horizon is three years and Globant pays no dividend, so none of the owner cash is distributed in the period; the model counts it as value economically available to owners. For a horizon of up to five years I look harder when the exit share passes about 85%. The Base is below that line, the Bull case (83.3% and 84.9%) just under it, and the Bear (69.4% and 74.9%) well under. A table default is not a justification for 11 times, so here is what it assumes. At a 10% required return, 11 times year-three cash is a 9.1% yield and implies growth of about 0.8% a year after 2029. Before that, Base cash grows 7.4% a year, from $125 million to $155 million, on revenue growth of 3.0% a year and a cash margin that rises from 5.1% to 5.8%. After 2029 the model assumes a franchise that lasts but no longer grows in real terms. The $85 million of annual investment is already deducted, so that small growth pays for itself. Debt stays at $422.0 million, refinanced, with $32 million of interest inside the cash. The Bear 8 times implies a decline of 2.2% a year, and the Bull 14 times growth of 2.7% a year. These multiples are my assumptions and not peer medians. The primary Bear is $14.72 (cash falling to $80 million, 8 times) and the Bull is $59.65 (cash rising to $210 million, 14 times). ## The secondary valuation: 10 times sustainable operating profit The second method starts from the current business and not from the 2029 endpoint. Sustainable operating profit is guided revenue of $2,445 million times a 14% adjusted margin, or $342.3 million, less $74.0 million of stock compensation and $50.0 million for recurring amortization of acquired intangibles and similar items: **$218.3 million**. That is 16% above the $187.5 million of IFRS operating profit I calculate for the twelve months to June 2026 (a 7.7% margin). The $50 million is my estimate, with a range of $35 million to $80 million. Ten times is an enterprise value of $2,183 million. After the $533.5 million of net claims above, equity is $1,649.5 million, or **$37.83** a share. I use 8 times on $170 million for the Bear ($18.96) and 13 times on $230 million for the Bull ($56.34). A 10 times multiple for a services business with flat revenue and an unfinished AI transition is my assumption. I could not assemble a same-date peer set with enterprise values and after-compensation cash multiples, so there is no peer discount or premium here. Lease treatment differs between the two methods on purpose. The cash method pays lease principal and interest in cash. The earnings method keeps right-of-use depreciation inside operating profit and leaves lease debt out of the bridge. ## The two methods on one basis The methods use different bases, one an equity cash flow and the other an enterprise profit, so a table that says they differ is not enough. Here both are converted to the same enterprise value and the same net claims, with the share count and bridge used throughout. | Base case | Cash method | Earnings method | | --- | --- | --- | | Equity value, $ million | 1,575.6 | 1,649.5 | | Net claims added, $ million | 533.5 | 533.5 | | Implied enterprise value, $ million | 2,109.1 | 2,183.0 | | Value per share | $36.14 | $37.83 | | Implied enterprise value to sustainable operating profit of 218.3 | 9.7 times | 10.0 times | | Equity value to year-three owner cash of 155 | 10.2 times | 10.6 times | | Equity value to today's owner cash of 125 | 12.6 times | 13.2 times | The earnings method values the same business 3.5% higher at enterprise level and 4.6% higher per share. In short, the cash method says Globant is worth 9.7 times operating profit and the earnings method says 10 times. The methods assume almost the same valuation, and the gap in value per share is $1.69. The cash method is also the stricter of the two in the Bear: it implies 6.9 times operating profit against the 8 times in the earnings method, so the Bear already values a shrinking Globant below a typical 8 times floor. In the Bull the cash method implies 13.6 times against 13 times. ## A price check against history The monthly chart shows the shares falling from peaks above $200 in 2024 and 2025 to $41.65. A fall that large does not make the old price a fair one: it reflected earnings and growth that no longer exist. I could not build a reconciled history of price to owner cash after compensation, leases and interest, so history carries no weight and no premium from the past is added. Peers help only on operations. In the quarter to June 2026 EPAM grew revenue 4.5% (3.4% organic in constant currency) with a non-GAAP operating margin of 16.4% and a GAAP margin of 10.8%. Cognizant grew 4.5% (4.1% in constant currency) and guides to an adjusted operating margin of 16.0% to 16.2% for 2026. Globant was flat with a 13.2% adjusted margin and a 3.2% IFRS margin. Accounting differs between IFRS and US GAAP, and so does the treatment of leases, so I do not turn this into a multiple. **The growth-adjusted check.** On the guided adjusted EPS of $5.75 to $6.15 the shares trade at 6.8 to 7.2 times, 7.0 at the midpoint. On the trailing IFRS EPS of $2.58 shown by a broker, the multiple is about 16.1. I do not divide either by a growth rate. FY2026 is not a clean next twelve months estimate, earnings and utilization are depressed and distorted by restructuring, and I have no synchronized forecast. A PEG of 2.16 on the broker page has an unknown method and gets no weight. The cash growth ratio (price to normalized owner cash, 14.5 times, divided by 7.4% a year of growth) is 1.95, but it measures a rebound from my own start point and proves nothing. ## How good is the cash growth? The three-year path starts from $125 million, my normalized estimate for October 2026, and not from the weak 2024 figure or from a reported number. | Owner cash, three years | Bear | Base | Bull | | --- | --- | --- | --- | | Start, $ million | 125 | 125 | 125 | | End, $ million | 80 | 155 | 210 | | Change over three years | -36.0% | +24.0% | +68.0% | | Annual growth, cash and cash per share | -13.8% | 7.4% | 18.9% | | Cash per share, start to end | 2.87 to 1.83 | 2.87 to 3.56 | 2.87 to 4.82 | | Endpoint revenue, $ million | 2,300 | 2,670 | 3,000 | | Endpoint cash margin | 3.5% | 5.8% | 7.0% | | Yield on today's model value of $1,816 million | 4.4% | 8.5% | 11.6% | The starting yield is 6.9%, and the twelve months to June 2026 gives about 7.0%. The Base growth comes from modest revenue growth and better utilization lifting operating cash flow, with working capital normalized, investment and compensation rising, interest rising with refinancing and minority holders taking more. It assumes no acquisitions and no pause in investment. If the share count grew 2% a year instead, Base cash per share would grow 5.3% a year, not 7.4%, but I do not charge that on top of the cash cost of compensation. Reliability of these estimates is low: they are my scenarios and not a consensus, and I could not find a synchronized analyst series for cash flow. With endpoint cash 20% lower ($124 million) the annual change is -0.3% and the yield 6.8%. With cash 40% lower ($93 million) it is -9.4% and 5.1%. ## Triangulating the values | Value per share today | Bear | Base | Bull | Weight | | --- | --- | --- | --- | --- | | Owner cash with an exit value | $14.72 | $36.14 | $59.65 | 65% | | Sustainable operating profit times a multiple | $18.96 | $37.83 | $56.34 | 35% | | Triangulated value | $16.20 | $36.73 | $58.49 | 100% | The Base methods differ by 4.6% (difference over midpoint), far under the 20% warning level, so no decomposition of the gap is needed and the table above already shows where it comes from. That agreement earns no confidence bonus, because both methods depend on the same recovery of the same services franchise. I gave the cash method 65% because what shareholders can actually receive is the binding constraint, and the earnings method 35% because it checks whether the severe cash deductions undervalue earning power. The weights are my judgment. Equal weights give $36.98 and the cash method alone gives $36.14; neither clears the 45% margin of safety. The weighted value is a point inside a range and does not narrow it. The Base range is $36.14 to $37.83, and the full range from Bear to Bull is $16.20 to $58.49. Choosing one method over the other does not change the decision. Both Base values sit below $41.65, and only the Bull case rises above it. ## What the price already assumes | Question at $41.65, one unknown at a time | Answer | | --- | --- | | Year-three owner cash needed at 11 times (cash method alone) | $181.7 million, 17.2% above Base, 13.3% a year from $125 million | | Same, in the weighted model | $191.6 million | | Exit multiple needed on Base cash (cash method alone) | 13.06 times | | Same, in the weighted model | about 13.8 times | | Sustainable operating profit needed at 10 times (earnings method alone) | $234.9 million, a 9.6% margin against 8.9% in the Base | Revenue does not have to explode. The price needs a better retained margin, better cash conversion and a longer life for the cash after 2029 than I assume, or some combination of them. These equations apply to one method at a time. Revenue, margin, working capital, claims and the multiple can offset each other, so this is a decomposition of the price and not a statement of what other investors believe. ## Bear, Base and Bull | | Bear | Base | Bull | | --- | --- | --- | --- | | Owner cash, years 1 / 2 / 3, $ million | 90 / 85 / 80 | 125 / 140 / 155 | 150 / 180 / 210 | | Exit multiple, cash method | 8 times | 11 times | 14 times | | Sustainable operating profit / multiple | $170m / 8 times | $218.3m / 10 times | $230m / 13 times | | Value per share today | $16.20 | $36.73 | $58.49 | | Against $41.65 | -61.1% | -11.8% | +40.4% | | Modeled annual return at $41.65, cash method | -23.4% | 4.7% | 24.5% | The Bear is persistent price and volume pressure, weak utilization and a smaller franchise. Both a falling cash path and an 8 times multiple are justified only because the weakness lasts beyond the explicit period; if it were temporary, I would test the multiple, not stack two pessimistic assumptions. The Bull needs growth to outrun price concessions and a lasting improvement in cash conversion, with higher investment and compensation still paid for. Interest, claims, share count and the discount rate are the same in the three cases. The returns are modeled branch outcomes with an economic realization convention, not dividends and not probabilities. ## Sensitivity and the case that flips the verdict | Weighted value, Base | 9 times | 11 times | 13 times | | --- | --- | --- | --- | | Year-three cash $125m | $29.90 | $32.70 | $35.50 | | Year-three cash $155m | $33.26 | $36.73 | $40.20 | | Year-three cash $185m | $36.62 | $40.76 | $44.91 | | Discount rate for the cash method | Cash method | Weighted value | | --- | --- | --- | | 8% | $38.10 | $38.01 | | 10% | $36.14 | $36.73 | | 12% | $34.31 | $35.54 | | Estimate changed | Range tested | Weighted value | Entry at 45% MOS | | --- | --- | --- | --- | | Year-three owner cash | $125m to $185m at 11 times | $32.70 to $40.76 | $17.98 to $22.42 | | Exit multiple | 9 to 13 times at $155m | $33.26 to $40.20 | $18.29 to $22.11 | | Acquisition claims | $148m to $80m | $36.09 to $37.65 | $19.85 to $20.71 | | Operating cash reserve | $150m to $75m | $35.58 to $37.30 | $19.57 to $20.52 | | Recurring adjustment in operating profit | $80m to $35m | $34.32 to $37.93 | $18.88 to $20.86 | | Refinancing rate | +300bp from May 2028 | $35.34 | $19.44 | One reasonable change in one input moves the value by a few dollars at most and never creates a Buy at $41.65. The entry price that gives a 45% margin of safety runs from about $18 to $22 across these tests. Changing only year-three cash to $140 million to $170 million and the multiple to 10 to 12 times gives an assumption-sensitive band of **$18.23 to $22.36**. The most favorable corner of the grid ($185 million, 13 times) gives $44.91, above the price, but a 45% margin on it is $24.70, still far below. A better operating outcome can lift the value above $41.65, as the Bull shows, so the conclusion is a conditional Hold and not a statement that the shares must fall. The band applies only while the operating and capital inputs stay valid; a refreshed capital bridge replaces it. ## How the buy price is set Confidence C calls for a margin of safety of 35% to 50%. I start at 40% and add 5 points for the part of value that sits beyond the modeled horizon: the exit value is 78.7% of the operating value, and that reliance on 2029 and after is charged here once, and nowhere else. The Bear scenario covers a different risk, a franchise that shrinks. The confidence score covers shared drivers and the wide spread of outcomes. Near-term demand, AI price pressure and refinancing are in the cash scenarios and the interest line, so they carry no extra charge. The maximum buy price is $36.73 x (1 - 45%) = **$20.20**. It is a decision reference, not an order to place. Three numbers answer different questions: what the shares are worth ($36.73), the annual return from buying at $41.65 and following the Base cash path (4.7%), and the price that would give a 12% return on that path ($34.31, cash method only). The 12% is an illustration and not a requirement I know you have, and it is not stacked on the margin of safety. At $41.65 the actual margin versus Base is -13.4%, so the shares do not qualify. ## The dividend Globant pays no dividend and its annual report describes a policy of retaining funds for operations and expansion, so the modeled yield is 0%. This is a capital appreciation and turnaround position, not an income holding. The buyback authorization is not a recurring payment: it improves per-share economics only after its cash cost, the replacement of compensation shares, liquidity needs and acquisition payments are covered, and a debt-funded repurchase around the 2028 maturity deserves a hard look. ## Reasons to own - **Recovery matters at this size.** A small improvement in revenue and retained cash margin changes the per-share result a lot, and the Bull case of $58.49 shows it. - **Enterprises still need implementation.** Integration, data, security and process redesign around AI remain work to buy even if raw coding gets cheaper. That is not a promise about pricing power. - **Existing relationships shorten adoption.** Selling outcome-based projects to current clients is easier than finding new ones, provided the new contracts earn a decent cash return. - **Cash flow has improved.** First half free cash flow was $48.7 million against -$8.6 million, and my bridge still gives credit for the improvement after leases, interest and compensation. - **Capital return can help.** Purchases well below my value of $36.73 would be constructive, with the $125 million authorization as the option and not the plan. ## Main risks - **AI price compression.** Fewer billable hours and client price savings can outweigh Globant's productivity gain. - **A weak demand cycle.** Postponed projects, a smaller base of large accounts and lower utilization. Accounts above $1 million fell to 331 and the ten largest customers are 30.6% of revenue. - **Adjusted profit does the talking.** Optimization charges recurred in 2025 and 2026, and stock compensation and amortization are large next to operating cash. - **Working capital swings.** Collection timing and payroll can flatter or hurt one quarter or one trailing year. - **Debt and the 2028 refinancing.** Higher rates, tighter covenants or restricted access to credit. A rate 300bp higher is worth about $1.39 a share in my weighted Base. - **Acquisition and minority claims.** Earn-outs and purchase options can consume common equity value, and I cannot split them with certainty. - **Goodwill.** Book equity overstates what the assets could earn or fetch. - **The chart as a false bargain.** An old multiple reflected economics that may no longer exist. ## Management execution Execution and forecasting should be judged separately, and the second is weaker. Globant opened 2025 with revenue guidance of $2,635 million to $2,705 million and delivered $2,454.9 million, 8.1% below the midpoint of $2,670 million. It opened 2026 with adjusted EPS guidance centered on $6.30 and cut it to $5.95 in August. It has booked business optimization charges in both years, and I count those as a recurring cost until the charges stop, not as a one-off. On the other side, the diluted share count fell 2.4% in a year and first half cash flow improved. The decision to keep cash and pay no dividend suits a business in transition, but its acquisition record does not show that purchased goodwill earns more than its cost. I score management 5/10 inside business quality, with mixed forecasting and mixed allocation, and apply no separate premium to the multiple. ## Stock Analyza scorecard | Business quality | Points | Maximum | | --- | --- | --- | | Moat and pricing power | 12 | 20 | | Economic return versus cost of capital | 4 | 20 | | Balance sheet and resilience | 11 | 15 | | Earnings and cash quality | 9 | 15 | | Growth and reinvestment runway | 7 | 15 | | Management and allocation | 5 | 10 | | Dilution, governance, minority alignment | 3 | 5 | | Total | 51 | 100 | On the scorecard for non-financial companies Globant scores 4/12, a standard company: 0/2 for return on capital because I could not build a reliable five-year series, 2/2 for five-year revenue growth ($814.1 million in 2020 to $2,454.9 million in 2025, 24.7% a year, though acquisitions inflate it), 0/2 for five-year cash per share, 1/1 for balance sheet capacity, 0/1 for exceptional unit economics, 1/2 for repeat relationships and 0/2 for a runway of high-return reinvestment. The valuation score is 38/100: 15/40 for the discount to value, 2/20 for protection against the Bear, 8/15 for agreement between methods, 3/10 for history, 7/10 for what the price implies and 3/5 for data quality. Business quality and valuation are separate numbers and are never blended. ## Confidence score | Component | Points of 5 | | --- | --- | | Data quality and provenance | 3 | | Business predictability | 3 | | Valuation robustness | 3 | | Accounting transparency | 3 | | Scenario dispersion | 1 | | Total, grade C (13 to 17) | 13 of 25 | A complete set of calculations does not turn estimates into reported facts. The latest complete annual cash flow statement is 2025 and the newest operating figures are the second quarter of 2026. To reach grade B I would need a current reconciliation of acquisition claims, several quarters of confirmed owner cash, a clearer cash bridge for the AI platform and a narrower spread of outcomes. ## What I would watch from here ### Supports the thesis - Organic revenue growth above 3% for two quarters across several clients, and accounts above $1 million back above 340. - Adjusted operating margin of at least 14.5% with clean conversion to cash, and rolling owner cash above $150 million without a pause in investment. - The 2028 debt refinanced on manageable terms, and buybacks funded from cash, below my value, with the share count falling. ### Watch closely - Revenue between -2% and +3%. Today's flat quarter is here. - Adjusted margin of 13.0% to 14.5% (second quarter: 13.2%), accounts above $1 million between 325 and 340 (today 331) and top ten concentration of 30% to 33% (today 30.6%). - Owner cash of $100 million to $150 million (my estimate is $125 million to $128 million), and no refinancing decision by late 2027. Guidance credibility already sits at the yellow to red border after the August cut. ### Breaks the thesis - Organic revenue below -2% for two quarters, or adjusted margin below 13% for good. - Rolling owner cash below $100 million, or accounts above $1 million below 325 and falling. - Covenant pressure, punitive refinancing or an equity raise, or buybacks funded with debt while cash and claims worsen. ## The most important thesis breakers - **Owner cash below $100 million with client erosion.** A shrinking customer base and a thinner cash margin together would put the Bear at the center. - **Financing.** A refinancing of the 2028 debt on punishing terms, or an equity raise, would rewrite interest, share count and the exit value. - **AI savings going to clients.** Falling cash per share while adjusted EPS stays steady means the productivity gain is not being kept. ## Stress test and independent check I rebuilt the model from its stated inputs with separate code and reproduced: the 2025 owner cash of $79.8 million and the rolled-forward $127.9 million; the primary values of $14.72, $36.14 and $59.65; the secondary values of $18.96, $37.83 and $56.34; the weighted values of $16.20, $36.73 and $58.49 and the 4.6% divergence; the $20.20 buy price and the $18.23 to $22.36 band; the three annual returns; the grids and tables above; the reverse results; and the three-year cash growth rates. Arithmetic was correct throughout. From Globant's releases I confirmed the second quarter and first quarter 2026 figures (revenue, operating profit, cash flow, capital expenditure, share counts, balance sheet lines, customer counts, Glob.AI, the buyback), the FY2025 results, and the guidance given in February 2025, February 2026, May and August 2026. EPAM and Cognizant figures come from their second quarter releases. The reference price is the October 20, 2026 close of $41.65 shown by stockanalysis.com, which was $40.14 a day earlier and $41.55 after hours. A broker quote early on October 21, before the US market opened, showed $41.63, a 0.05% difference. The verdict is the same at either. I could not re-read some inputs, so they come from the supplied analysis: the audited 2023 to 2025 cash flow lines, the borrowing, lease and liquidity notes and the covenant terms in the 20-F, the minority and acquisition liability detail, the prior-year top ten customer share of 29.3%, the first half 2025 comparatives, and the sustainable operating profit inputs of stock compensation and amortization. Where I could cross-check part of a chain, such as the first half free cash flow of $48.7 million (first quarter $36.1 million plus second quarter $12.6 million), it matched. The newest version of my method asks for three disclosures that earlier versions did not: what each multiple assumes, the two methods on one basis, and cash before and after financing for a leveraged company. All three are above, and none changed the value. The only changes to the figures came from using the verified closing price instead of the broker quote. | Test | Weighted value | Entry at 45% MOS | Effect | | --- | --- | --- | --- | | Endpoint owner cash 20% lower ($124m) | $32.56 | $17.91 | Hold, margin unmet | | Endpoint owner cash 40% lower ($93m) | $28.40 | $15.62 | Downside to the center is material | | Bear exit multiple 7 times instead of 8 | $15.31 | $8.42 | Bear falls; Base unchanged | | Bear exit multiple 6 times instead of 8 | $14.41 | $7.93 | Bear falls; Base unchanged | | Refinancing +300bp from May 2028 | $35.34 | $19.44 | Hold, margin unmet | | Favorable corner: $185m and 13 times | $44.91 | $24.70 | Above the price, but no margin of safety | | Bull cash at an 8% discount rate | $60.61 | $33.33 | Buy economics only under strong execution | The strongest case against Hold is that acquired intangible amortization costs less in cash than reported, demand recovers and Globant keeps the AI productivity gain. Then owner cash reaches $210 million at 14 times, the weighted Bull is $58.49 and the shares are 40.4% undervalued. That is a serious case and I have not dismissed it. The strongest case against owning at all is that AI tools permanently shrink profitable engineering demand. Then cash falls to $80 million at 8 times, the weighted value is $16.20, 61.1% below the price, and a refinancing problem could make it worse. My Base grants a recovery from the historical cash level and does not assume the Bear is certain. My corrections were not one-directional either: the earnings method deliberately challenges an excessively harsh cash-only view. Main data limits: I do not have a current share count, a split of the $148.0 million of other financial liabilities, covenant headroom, a synchronized cash flow consensus or a same-date peer valuation. These are why Confidence is C. ## Final verdict: hold at $41.65 Globant can recover, and the Bull case is not a fantasy. But the price pays for most of that recovery already, because the gap between adjusted profit of about $259 million and owner cash of about $125 million is made mostly of investment, stock compensation, leases, interest and minority holders that shareholders carry. Customer breadth is narrowing, guidance has been cut, and a $422 million refinancing sits inside the horizon. A new purchase is not supported by this model at a price above $20.20. **Verdict: HOLD at $41.65, watchlist for new money. Triangulated Base value $36.73, downside 11.8%, Confidence C (13/25), required margin of safety 45%, maximum buy price $20.20, assumption-sensitive band $18.23 to $22.36.** For holders this is a review, not an instruction to sell, and portfolio circumstances differ. For new money it means waiting for a lower price with the thesis intact, or for several quarters of owner cash above $150 million without cuts to investment. The cheap number is the adjusted EPS. The number that matters is the $125 million a year, and the price already asks for close to $182 million of it by 2029. ## More analyses on Patreon This analysis is free. Premium members get every in-depth analysis, the full archive and a vote on which stocks get analyzed next: https://www.patreon.com/stockanalyza ## Sources Primary sources include Globant's [second quarter 2026 results on SEC Exhibit 99.1](https://www.sec.gov/Archives/edgar/data/1557860/000110465926095916/tm2623064d1_ex99-1.htm) and the [investor relations version](https://investors.globant.com/2026-08-13-Globant-Reports-2026-Second-Quarter-Financial-Results), the [first quarter 2026 results](https://investors.globant.com/2026-05-14-Globant-Reports-2026-First-Quarter-Financial-Results), the [2025 fourth quarter and full year results](https://investors.globant.com/2026-02-26-Globant-Reports-2025-Fourth-Quarter-Financial-Results), the [audited Form 20-F for 2025](https://www.sec.gov/Archives/edgar/data/1557860/000162828026012910/glob-20251231.htm), the [2024 results with the initial 2025 guidance](https://investors.globant.com/2025-02-20-Globant-Reports-2024-Fourth-Quarter-and-Full-Year-Financial-Results), the [2020 results](https://investors.globant.com/2021-02-18-Globant-Reports-2020-Fourth-Quarter-and-Full-Year-Financial-Results) for the five-year revenue start and the [share repurchase authorization](https://investors.globant.com/2026-05-18-Globant-launches-new-125-million-share-repurchase-program). Peer figures come from [EPAM's second quarter 2026 release](https://www.sec.gov/Archives/edgar/data/0001352010/000135201026000043/exhibit99_q2x2026.htm) and [Cognizant's second quarter 2026 release](https://investors.cognizant.com/news-and-events/news/news-details/2026/Cognizant-Reports-Second-Quarter-2026-Results/default.aspx). The reference market price is $41.65, the October 20, 2026 regular session close, shown on the [stockanalysis.com quote page](https://stockanalysis.com/stocks/glob/). Per share values use 43.6 million diluted shares, the company's 2026 guidance assumption. The owner cash path, the sustainable operating profit, the multiples, the 10% discount rate, the acquisition claim of $120 million, the $100 million operating cash reserve and the minority cash share are analytical estimates, not company guidance. **Disclaimer.** This analysis is provided solely for research and educational purposes. It is not personalized investment, financial, legal or tax advice. Values are estimates, not forecasts or trading instructions. Globant is exposed to pricing pressure from AI tools, weak client demand, customer concentration, working capital swings, debt maturing in 2028, acquisition related liabilities and goodwill impairment. Investors can lose part or all of their invested capital. Method: Master Investment Analysis Framework v3.5. Approach: three-year owner cash model with a discounted exit value, cross-checked with normalized operating profit and a common-basis comparison. Confidence: C/13. Data checks: passed with warnings. Conditional research only. Open data gaps: current share count, acquisition liability detail and covenant headroom.