Microsoft (MSFT) — DCF Model Summary
Published · Data as of 2026-10-19 close; filings through 2026-10-20. Figures are as of the dates shown and will not be updated.
How this report was made. An AI research agent (Claude Code running Anthropic's open-source finance skills, adapted by bests.one) produced this report on 2026-10-20 using only free data: SEC EDGAR filings and XBRL, FRED, and unofficial Yahoo Finance prices. Web-sourced figures (analyst consensus, call commentary) are tagged
[WEB]with provider, date and link; a separate checking pass confirmed all 6 against the original pages. A separate AI checking pass compared 17 of 17 figures with the original sources (SEC filings, FRED); all matched except where noted. Reviewed by bests.one on 2026-10-20. The run took about 4 minutes and cost about $0.82 in compute. Read the method: How we test.Educational sample, not investment advice. Nothing here is a recommendation to buy, sell or hold any security. bests.one is independent and not affiliated with NVIDIA, Microsoft, Manhattan Associates, Anthropic, or any data provider mentioned.
Read this first. The scenario values below come from our own assumptions, mainly the capex path and the terminal value. They are not consensus and not a price target. The perpetuity-growth and exit-multiple methods give very different answers ($348 vs $522 in the base case), which is the point of the exercise: a DCF shows which assumptions matter, not what a stock is "worth".
Analysis date: 2026-10-20 · Price: $533.52 (Yahoo Finance chart data, unofficial; close of 2026-10-19 16:00 exchange local; retrieved 2026-10-20)
Units: $ in millions; shares in millions; per-share in $. Fiscal year ends June 30 (FY2026 = 2025-07-01 to 2026-06-30).
Balance sheet date: 2026-06-30 (FY2026 10-K). The Q1 FY2027 10-Q (quarter to 2026-09-30) is not yet filed; MarketBeat estimates the report for 2026-10-28, unconfirmed by Microsoft [WEB: MarketBeat, retrieved 2026-10-20].
SEC-verified vs web-sourced. SEC-verified (us-free-finance: XBRL, filings, FRED, Yahoo prices): all historicals, balance sheet, shares, price, beta, risk-free rate, CPI. Web-sourced, each tagged [WEB], used only as cross-checks and never in the valuation: FY2027 consensus and verbal call guidance. The projections are my own assumptions, not consensus and not guidance.
Scenarios, not a recommendation. Research draft based on free public data; not investment advice.
1. Result summary
| Scenario | EV @ 2026-06-30 | Equity value @ 2026-10-20 | Value/share (perpetuity growth) | vs $533.52 | Value/share (exit-multiple cross-check) |
|---|---|---|---|---|---|
| Bear | 1,420,300 | 1,471,163 | $198 | -63% | $293 (12x EBITDA) |
| Base | 2,501,401 | 2,586,031 | $348 | -35% | $522 (15x EBITDA) |
| Bull | 3,604,600 | 3,723,689 | $500 | -6% | $770 (18x EBITDA) |
Key points:
- Under a 10.5% WACC (driven by a 5.09% 10-year Treasury yield), no scenario reaches the current price on perpetuity growth. Only the Bull case comes close.
- The two terminal methods diverge widely. In the Base case, the perpetuity-growth terminal value implies only 8.1x FY2036E EBITDA (6.4x Bear, 9.7x Bull). Microsoft trades at 19.8x LTM EBITDA today (comps snapshot). The 12–18x exit multiples are my assumptions, not peer-derived, and they are what produce the higher per-share values.
- The market price is consistent with some combination of: a lower discount rate (about 8.5% WACC with g = 3.5% gives about $500; see Table 1), higher long-run growth or margins than Base, or a terminal multiple well above what a 10.5% WACC and 3.5% growth support.
- Terminal value is 55–63% of EV. The result is highly sensitive to capex normalization (section 4).
2. Historical analysis (FY2022–FY2026, $M)
Source: get_financials annual, 10-K filings. Filing index links: FY2024 10-K filed 2024-07-30, FY2025 10-K filed 2025-07-30, FY2026 10-K filed 2026-07-29. FY2022 comes from the FY2024 10-K comparatives.
| FY2022 | FY2023 | FY2024 | FY2025 | FY2026 | |
|---|---|---|---|---|---|
| Revenue | 198,270 | 211,915 | 245,122 | 281,724 | 331,839 |
| growth | n/a | 6.9% | 15.7% | 14.9% | 17.8% |
| Gross margin | 68.4% | 68.9% | 69.8% | 68.8% | 67.9% |
| Operating income (EBIT) | 83,383 | 88,523 | 109,433 | 128,528 | 155,237 |
| EBIT margin | 42.1% | 41.8% | 44.6% | 45.6% | 46.8% |
D&A (cash-flow line, msft:DepreciationAmortizationAndOther) | 14,460 | 13,861 | 20,958 | 29,433 | 38,534 |
| D&A % revenue | 7.3% | 6.5% | 8.5% | 10.4% | 11.6% |
| Capex (cash PP&E purchases) | 23,886 | 28,107 | 44,477 | 64,551 | 115,948 |
| Capex % revenue | 12.0% | 13.3% | 18.1% | 22.9% | 34.9% |
| Capex / D&A | 1.65x | 2.03x | 2.12x | 2.19x | 3.01x |
| Cash from operations | 89,035 | 87,582 | 118,548 | 136,162 | 182,935 |
| FCF (CFO − capex, derived) | 65,149 | 59,475 | 74,071 | 71,611 | 66,987 |
| Stock-based comp | 7,502 | 9,611 | 10,734 | 11,974 | 12,405 |
| Effective tax rate | 13.1% | 19.0% | 18.2% | 17.6% | 19.4% |
| Net income | 72,738 | 72,361 | 88,136 | 101,832 | 133,749 |
| Diluted EPS ($) | 9.65 | 9.68 | 11.80 | 13.64 | 17.95 |
| Net PP&E (year-end) | 74,398 | 95,641 | 135,591 | 204,966 | 313,076 |
Observations:
- Revenue growth accelerated to 17.8% in FY2026. The Q4 FY2026 press release (8-K EX-99.1, 2026-07-29) reports Q4 revenue of $90.0B (+18%), Microsoft Cloud revenue of $59.3B (+27%), Azure and other cloud services +43%, and commercial remaining performance obligation of $678B (+84%).
- EBIT margin expanded about 4.7 points over four years even as D&A rose from 7% to 12% of revenue. Capex has gone from 12% to 35% of revenue. FCF has been flat to down since FY2024 despite CFO growing 54% over the same period.
- FY2026 net income includes OpenAI-related gains of $4,963M and a $3.2B Anthropic gain in Q4, all below the EBIT line. EBIT is not affected.
- D&A restatement: the FY2026 10-K shows FY2024 D&A of 20,958 (originally 22,287) and FY2025 D&A of 29,433 (originally 34,153). I used the latest filing, per the restatement rule. Separately, the comps snapshot sums quarterly D&A to 43,448 for LTM, against 38,534 on the annual line. I did not reconcile the difference and used the annual line.
- Working capital (the free tools do not normalize "other current assets" or the other current-liability lines, so I took them from the 10-K balance sheet in the 8-K EX-99.1; only two years are available this way):
| Operating NWC ($M) | FY2025 | FY2026 |
|---|---|---|
| AR + inventory + other current assets | 96,566 | 130,867 |
| AP + accrued comp + ST income taxes + ST unearned revenue + other current liabilities | 138,219 | 159,598 |
| NWC | -41,653 (-14.8% of revenue) | -28,731 (-8.7% of revenue) |
Microsoft runs negative working capital because customers prepay (unearned revenue). Other current assets nearly doubled in FY2026, from 25,723 to 48,594; I did not investigate why.
3. Assumptions and rationale
| Driver | Bear | Base | Bull | Rationale |
|---|---|---|---|---|
| Revenue growth FY27→FY36 | 13, 10, 8, 7, 6, 5.5, 5, 4.5, 4, 3.5% | 17, 15, 13.5, 12, 10.5, 9, 8, 7, 6, 5% | 19, 18, 16.5, 15, 13, 11.5, 10, 8.5, 7, 6% | FY26 was 17.8% and Q4 was 18%. Commercial RPO of $678B (+84%) supports near-term visibility. Base FY27 revenue of $388.3B is within 1% of the web consensus ($390.7B). Growth fades toward nominal GDP. |
| EBIT margin | 43% → 35% | 45.5% → 42% | 47% → 48% → 47% | FY26 was 46.8%. Depreciation from the $313B PP&E base (up 53% in FY26) will weigh on margins. Base fades about 5 points; Bear assumes AI infrastructure returns disappoint; Bull holds margins. |
| Capex % revenue (cash PP&E purchases) | 38% → 17.5% | 38% → 15.5% | 38% → 14.5% | FY26 was 34.9%. Management said on the call that FY27 capex will grow year over year [WEB] (section 8). I set FY27 at 38% (about $148B Base). That is my assumption, not guidance. |
| D&A | 17% of beginning net PP&E (all cases) | FY26 actual was 18.8%. I used 17% because Microsoft extends the life of data centers and buildings from 15 to 25 years from FY27 [WEB: CFO Dive]. PP&E roll-forward: end = beginning + capex − D&A. It ignores non-cash finance-lease additions and intangibles. | ||
| Tax rate on EBIT | 21% | 19.5% | 18% | FY26 effective rate was 19.4%; 3-year average 18.4%. |
| Change in NWC | 0% of Δrevenue | inflow of 5% of Δrevenue | inflow of 8% of Δrevenue | NWC is negative and was -8.7% of revenue in FY26 (-14.8% in FY25). A modest inflow is assumed, and it is a small part of FCF. |
| SBC | Not added back | EBIT is after SBC ($12.4B in FY26) and I treat it as a real cost. | ||
| Terminal growth | 3.0% | 3.5% | 4.0% | The CPI index of 336.298 (Sep 2026) vs 324.245 (Sep 2025) is +3.7% year on year. The 10-year yield is 5.09%. Terminal growth is an assumption. |
| Exit EV/EBITDA (FY36E) | 12x | 15x | 18x | My assumption. Peer snapshot at LTM: GOOGL 23.5x, AMZN 16.8x, ORCL 11.2x (flagged understated), MSFT 19.8x. Not a vetted peer set (see gaps). |
The projection is 10 years, not 5, because capex intensity at 35% of revenue cannot be normalized inside 5 years. Terminal-year reinvestment is set so that net capex grows PP&E at g: capex(FY37) = D&A(FY37) + g × PP&E(FY36). This avoids assuming growth without the reinvestment needed for it.
4. Unlevered FCF projection ($M, Base case)
Full Bear/Base/Bull rows are in MSFT_DCF_projection.csv. UFCF = EBIT × (1 − t) + D&A − capex − ΔNWC.
| FY26A | FY27E | FY28E | FY29E | FY30E | FY31E | FY32E | FY33E | FY34E | FY35E | FY36E | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Revenue | 331,839 | 388,252 | 446,489 | 506,765 | 567,577 | 627,173 | 683,618 | 738,308 | 789,990 | 837,389 | 879,258 |
| growth | 17.8% | 17.0% | 15.0% | 13.5% | 12.0% | 10.5% | 9.0% | 8.0% | 7.0% | 6.0% | 5.0% |
| EBIT margin | 46.8% | 45.5% | 45.0% | 44.5% | 44.0% | 43.5% | 43.0% | 42.5% | 42.0% | 42.0% | 42.0% |
| D&A | 38,534 | 53,223 | 69,256 | 84,049 | 96,467 | 106,119 | 113,668 | 118,749 | 122,409 | 125,102 | 127,323 |
| Capex | 115,948 | 147,536 | 156,271 | 157,097 | 153,246 | 150,521 | 143,560 | 140,279 | 138,248 | 138,169 | 136,285 |
| Capex / D&A | 3.01x | 2.77x | 2.26x | 1.87x | 1.59x | 1.42x | 1.26x | 1.18x | 1.13x | 1.10x | 1.07x |
| UFCF | n/a | 50,715 | 77,637 | 111,501 | 147,298 | 178,198 | 209,565 | 233,799 | 253,841 | 272,424 | 290,409 |
| Net PP&E (end) | 313,076 | 407,389 | 494,404 | 567,453 | 624,231 | 668,634 | 698,526 | 720,055 | 735,894 | 748,961 | 757,923 |
Reference: FY26 reported FCF (CFO − capex) was 66,987. UFCF is not computed for FY26 because it would need an NWC and tax-shield treatment on the actual figures that I did not model.
Cross-check against the web consensus [WEB: Simply Wall St (analyst aggregate; underlying data: S&P Global Market Intelligence, per the page footer), 2026-08-01]: FY2027 revenue of $390.7B (48 analysts) versus my Base of $388.3B. Consensus FY2027 EPS is $19.64. I did not build an EPS bridge, so there is no EPS comparison.
Scenario summary (FY2036E): Bear revenue 629,570, EBIT margin 35%, UFCF 176,236. Bull revenue 1,064,745, EBIT margin 47%, UFCF 394,072.
5. WACC build
| Input | Value | Source |
|---|---|---|
| Risk-free rate | 5.09% | FRED DGS10, observation date 2026-10-16 (series) |
| Beta (raw) | 1.1352 | OLS, 104 weekly adjusted-close returns vs SPY, 2024-10-21 to 2026-10-19; R² 0.309; standard error 0.168 |
| Beta (Blume-adjusted, used) | 1.0906 | (2/3 × raw + 1/3 × 1). The raw beta's 95% interval is roughly 0.80–1.47. |
| Equity risk premium | 5.0% | Assumption (tool default), not a data item |
| Cost of equity | 10.54% | 5.09% + 1.0906 × 5.0%. With raw beta: 10.77%. |
| Pre-tax cost of debt | 7.31% | FY2026 interest expense of 3,051 / average book debt of 41,723. This is likely overstated because the interest expense probably includes finance-lease interest while the debt base excludes leases. Impact is small at a 1% debt weight. |
| Tax rate | 19.4% (used 19.5%) | FY2026 effective rate |
| Weights | E 98.99% / D 1.01% | Market cap $3,961,677M (price × 7,425.5M cover shares as of 2026-07-23) and book debt of $40,294M excluding leases. Including finance leases, the debt weight would be about 2.6% and WACC about 0.1 point lower. |
| WACC | 10.50% |
The 5.09% risk-free rate is the tool's FRED value and is high relative to my own recollection of recent rates. I did not override it. Table 3 shows how much it matters.
6. Terminal value and EV-to-equity bridge
| ($M) | Bear | Base | Bull |
|---|---|---|---|
| PV of FY27–FY36 UFCF (mid-year, valuation date 2026-06-30) | 635,829 | 1,012,022 | 1,346,504 |
| Terminal growth | 3.0% | 3.5% | 4.0% |
| FY37 normalized UFCF | 159,539 | 282,693 | 397,968 |
| Terminal value (Gordon) | 2,128,339 | 4,040,815 | 6,126,410 |
| PV of terminal value | 784,472 | 1,489,379 | 2,258,096 |
| Enterprise value | 1,420,300 | 2,501,401 | 3,604,600 |
| TV as % of EV | 55.2% | 59.5% | 62.6% |
| Implied TV / FY36E EBITDA | 6.4x | 8.1x | 9.7x |
| Cross-check: exit multiple | 12x | 15x | 18x |
| Exit-multiple TV | 3,992,211 | 7,449,178 | 11,406,886 |
| Exit-method EV | 2,107,293 | 3,757,669 | 5,550,899 |
Bridge (all scenarios, balance sheet at 2026-06-30, FY2026 10-K):
| Item | $M |
|---|---|
| Enterprise value | per scenario |
| − Debt (current 9,227 + long-term 31,067) | −40,294 |
− Finance lease liabilities (FinanceLeaseLiability, 10-K filed 2026-07-29) | −66,594 |
| + Cash and cash equivalents | +20,935 |
| + Short-term investments | +55,908 |
| + Equity and other investments (book value; includes OpenAI and Anthropic stakes) | +36,348 |
| Net: equity = EV + 6,303 | +6,303 |
| = Equity value at 2026-06-30 | Bear 1,426,603 / Base 2,507,704 / Bull 3,610,903 |
| × (1 + Ke)^(112/365) = 1.0316, to roll forward to 2026-10-20 | |
| = Equity value at 2026-10-20 | Bear 1,471,163 / Base 2,586,031 / Bull 3,723,689 |
| ÷ diluted shares 7,441.5M (cover shares 7,425.5M + 16M dilution, the Q4 FY26 diluted-minus-basic gap) | |
| Value per share | $198 / $348 / $500 |
Notes on the bridge:
- Operating lease liabilities are not deducted, because EBIT is after operating-lease cost. Finance leases are deducted as debt-like. Long-term and other marketable securities beyond the "equity and other investments" line are not captured.
- "Equity and other investments" at book value is a judgment call. Excluding it would lower value per share by about $5 (36,348 × 1.03 / 7,441.5). The OpenAI stake's market value is not available from free data.
- The roll-forward from 2026-06-30 assumes no cash-flow leakage (dividends of roughly $0.91 per quarter and buybacks are ignored) and does not add the July–October 2026 cash flow separately; the roll-forward at Ke is the proxy.
- Perpetuity-growth value per share vs price: Bear -63%, Base -35%, Bull -6%.
7. Sensitivity tables (Base case, value per share in $; vs $533.52)
Table 1: WACC × terminal growth
| WACC \ g | 2.5% | 3.0% | 3.5% | 4.0% | 4.5% |
|---|---|---|---|---|---|
| 8.50% | 447 | 471 | 500 | 536 | 580 |
| 9.50% | 377 | 393 | 411 | 433 | 459 |
| 10.50% | 324 | 335 | 348 | 362 | 378 |
| 11.50% | 284 | 292 | 300 | 310 | 321 |
| 12.50% | 252 | 257 | 264 | 270 | 278 |
(Cost of equity is moved with WACC when rolling forward to 2026-10-20.)
Table 2: Revenue growth shift (every year, percentage points) × EBIT margin shift (every year, percentage points), at WACC 10.5% and g 3.5%
| Growth Δ \ Margin Δ | -6 pts | -3 pts | 0 | +3 pts | +6 pts |
|---|---|---|---|---|---|
| -4 pts | 218 | 240 | 261 | 283 | 305 |
| -2 pts | 251 | 276 | 301 | 326 | 351 |
| 0 | 290 | 319 | 348 | 376 | 405 |
| +2 pts | 335 | 368 | 402 | 435 | 468 |
| +4 pts | 388 | 426 | 464 | 503 | 541 |
Growth shifts apply to each year's growth rate (a −4 pt shift on 17% gives 13%, and so on). Terminal growth stays at 3.5%.
Table 3: Beta × risk-free rate (ERP 5.0%, g 3.5%)
| Beta \ Rf | 4.0% | 4.5% | 5.09% | 5.5% | 6.0% |
|---|---|---|---|---|---|
| 0.90 | 503 | 454 | 407 | 379 | 350 |
| 1.00 | 454 | 413 | 374 | 350 | 324 |
| 1.09 (Blume) | 417 | 382 | 348 | 327 | 304 |
| 1.135 (raw) | 401 | 368 | 336 | 316 | 295 |
| 1.30 | 350 | 324 | 299 | 283 | 266 |
Only the top-left corner of Table 3 (low beta and low risk-free rate) and the high-growth/high-margin corner of Table 2 reach the current price. Terminal growth in Table 1 at 8.5% WACC and 4.5% reaches $580.
8. Web supplements (labeled; not used in valuation)
| Item | Value | Tag | Status |
|---|---|---|---|
| FY2027 revenue consensus | $390.7B (48 analysts; up from $385.0B pre-results) | [WEB: Simply Wall St aggregate of S&P Global Market Intelligence data, 2026-08-01, https://simplywall.st/stocks/us/software/nasdaq-msft/microsoft/news/earnings-update-microsoft-corporation-nasdaqmsft-just-report] | Page fetched. Single source. |
| FY2027 EPS consensus | $19.64 | same | Page fetched. Basis (GAAP or adjusted) not stated. |
| Calendar-2026 capex and accounting change | Calendar-2026 capex outlook cut to about $175B from $190B because more data-center leases move from finance to operating leases; useful life for data centers and buildings 15 → 25 years from FY27; "minimal benefit" to FY27 operating income | [WEB: CFO Dive, 2026-07-30, citing CFO Amy Hood on the earnings call — verbal guidance, not in the filing, https://www.cfodive.com/news/microsoft-holds-line-ai-spending-plans/826648/] | Page fetched. Hood: capex expectations "remain unchanged" outside the useful-life effect. |
| FY2027 capex "will grow year over year"; Q1 FY27 capex above $50B; Q1 revenue guide $89.9–91.0B; Visible Alpha FY27 revenue of $391B and EPS of $19.66 | search snippets (Nasdaq/MarketBeat, S&P Global) | Search snippet only, not verified. Not used. | |
| Reported FY27 capex of $255–260B (Sentisense) | search snippet | Contradicted by other sources and not verified. Not used. |
Implication for this model: the useful-life change moves future data-center leases from finance leases (counted in capex) to operating leases (not counted). That makes my "cash PP&E capex" definition less comparable to prior years going forward (see gaps).
9. Sources
SEC filings (all via us-free-finance; XBRL companyfacts retrieved 2026-10-20):
- FY2026 10-K, filed 2026-07-29, accession 0001193125-26-323660: index, document. Provides FY2026 income, cash-flow and balance sheet lines;
LongTermInvestments36,348;FinanceLeaseLiability66,594; shares on cover as of 2026-07-23. - FY2025 10-K, filed 2025-07-30, accession 0000950170-25-100235: index.
- FY2024 10-K, filed 2024-07-30, accession 0000950170-24-087843: index.
- FY2023 balance sheet: 10-K filed 2023-07-27, accession 0000950170-23-035122: index.
- 8-K, item 2.02, filed 2026-07-29, accession 0001193125-26-323632, EX-99.1: document. Used for the working-capital detail and Q4 highlights. The release's "Business Outlook" section defers guidance to the call, so there is no outlook text in the filing.
FRED / Yahoo:
- 10-year Treasury yield, DGS10, 5.09% as of 2026-10-16: https://fred.stlouisfed.org/series/DGS10
- CPI-U, CPIAUCSL, latest 336.298 (2026-09-01): https://fred.stlouisfed.org/series/CPIAUCSL
- Yahoo Finance chart data (unofficial), retrieved 2026-10-20: MSFT price $533.52 (2026-10-19 close), 52-week range $349.20–$553.72; weekly MSFT and SPY history for beta; GOOGL, AMZN, ORCL prices for the comps snapshot.
Web (access date 2026-10-20): Simply Wall St (2026-08-01) and CFO Dive (2026-07-30), as tagged in section 8.
10. Data gaps and caveats
- No consensus in free data beyond one web aggregate. One FY2027 revenue and EPS figure was found; there are no FY2028+ estimates and no consensus capex, margin or free cash flow. The projections are my assumptions.
- No management guidance in the filings. The 8-K defers outlook to the call. FY2027 capex guidance was not found on a verified page; the only verified guidance is directional (calendar-2026 about $175B on the new accounting basis).
- Capex excludes finance-lease additions. The cash PP&E line (115,948 in FY26) excludes finance leases, which for Microsoft are large (liability of $66.6B at 2026-06-30, up from $60.2B at 2025-12-31). True economic capex is higher than my FCF assumes, and the new lease-classification policy changes what the line includes. I did not find the finance-lease addition figure for each year; FCF in the model could be overstated by that amount. I partly offset this by deducting the finance-lease liability in the bridge. This is a material limitation.
- Latest quarter not available. Q1 FY2027 (quarter ended 2026-09-30) is not yet reported. The balance sheet is as of 2026-06-30 and the valuation is rolled forward at the cost of equity.
- Peers not vetted. I ran
get_comps_snapshotfor GOOGL, AMZN and ORCL but did not runget_peer_candidatesor verify the peer set; these three are shown only as context for the exit-multiple assumption. Several peer EBITDA figures are flagged as understated (depreciation-only tags). No forward multiples were found. Exit multiples are assumptions. - OpenAI and Anthropic stakes. Carried at book value within "equity and other investments." Fair value, the commercial terms of the OpenAI relationship and the portion of RPO from OpenAI are not available from free data and are not modeled.
- Beta quality. R² of 0.31 and a standard error of 0.17 over 104 weekly observations. The 5.09% risk-free rate and the 5.0% ERP are inputs; changing them moves value materially (Table 3).
- D&A inconsistency. The annual D&A line was restated in the FY2026 10-K and differs from the summed quarterly D&A in the comps snapshot (38,534 vs 43,448 LTM).
- Segment build not done. Revenue is modeled in total, not by segment. Segment revenue (Productivity and Business Processes, Intelligent Cloud, More Personal Computing) is available in the 10-K but not modeled here.
- Operating NWC used the 10-K balance sheet from the 8-K exhibit for two years only; it is a small contributor to value.
- Quarterly working capital, stock-based dilution (RSUs) and the tax rate on foreign income are simplified. Diluted shares use the Q4 FY26 gap between diluted and basic weighted shares.
- Ratings and price targets are not available and are not used.
Research draft based on free public data (SEC EDGAR, FRED, Yahoo Finance and the web sources listed); not investment advice.
Download the data behind this note: MSFT_DCF_projection.csv (CSV).