Full Report

Bottom line

Accenture's three-year record is one of a slowdown that ended and a recovery that is real but modest. Revenue growth in local currency fell from 8% in FY2023 to just 2% in FY2024 as discretionary consulting demand stalled, then reaccelerated to 7% in FY2025 on AI-led reinvention work [1] [2]. Adjusted operating margin ground higher by roughly 10 basis points a year to 15.6%, and adjusted EPS compounded to $12.93 [3]. Free cash flow reached a record $10.9 billion [4].

Is that trajectory durable enough to support the forward estimates? On balance, yes — but the forward path is guided to deceleration, not acceleration. Management's own FY2026 outlook (as of June 18, 2026) is for 3%–4% local-currency growth, held back about a point by a shrinking U.S. federal business, with adjusted EPS of $13.78–$13.90 [5]. The reliable engine here is margin and cash, not top-line: a book-to-bill that has slipped below 1.2 and bookings that fell 2% year-over-year in the latest quarter are the reason a base case should not assume a return to high-single-digit growth [6].

The five-year picture

Sources: FY2023–FY2025 as reported — Q4 FY2025 press release, income statement and reconciliations [7] [8]; Q4 FY2024 [9]; Q4 FY2023 [10]. FY2026E per management guidance [11]; FY2027E is a modeled base case (assumptions below).

Source: Q4 and Full-Year FY2025 results — revenue and bookings [12]; adjusted margin, EPS and free cash flow [13].

Sources: reported FY2023–FY2025 [14]; FY2026E–FY2027E modeled (blue = actual, amber = estimate).

What drove the three-year record

The story is best told in local currency, which strips out a swing from a roughly 4% FX headwind in FY2023 to a positive tailwind by FY2025.

FY2023 — strong growth, weak optics. Revenue rose 8% in local currency but only 4% in dollars, and GAAP operating margin fell to 13.7% because Accenture booked a heavy $1.06 billion business-optimization (severance) charge, worth $1.28 of EPS [15]. On an adjusted basis the margin actually expanded 20 basis points to 15.4% and adjusted EPS grew 9% to $11.67 — the gap between the GAAP and adjusted lines is the single most important thing to understand about this company's reported numbers [16].

FY2024 — the air pocket. Growth collapsed to 2% local currency / 1% dollars as clients pulled back on discretionary consulting; managed-services and bookings held up (new bookings jumped 13% to $81.2 billion, a book-to-bill near 1.25) while revenue conversion lagged [17] [18]. A smaller $438 million optimization charge lifted GAAP margin optically by 110 basis points to 14.8%; adjusted margin inched up just 10 basis points to 15.5%, and adjusted EPS grew only 2% [19]. This is the year that best frames the durability question: Accenture kept selling work but could not convert it to revenue growth.

FY2025 — AI-led reacceleration. Revenue grew 7% in both dollars and local currency to $69.7 billion, the FX headwind having turned neutral [20]. The visible driver was AI: generative/"advanced" AI bookings nearly doubled to $5.9 billion and related revenue tripled to $2.7 billion, on the back of the $3 billion multi-year AI investment committed in FY2023 [21]. Adjusted operating margin rose 10 basis points to 15.6% and adjusted EPS grew 8% to $12.93; note, though, that total new bookings actually fell 1% to $80.6 billion — the growth reacceleration was not matched by a bookings reacceleration [22] [23].

Sources: FY2023–FY2025 reported GAAP and adjusted operating margins [24] [25] [26]; FY2026E per guidance [27], FY2027E modeled. The grey-vs-green wedge each year is the business-optimization charge.

The margin picture is Accenture's most durable feature: on an adjusted basis the operating margin has expanded every year for three years, at a deliberately steady ~10 basis points per annum — this is a management commitment, not an accident, and it is the backbone of the forward EPS estimate. The GAAP line is far noisier only because the severance charges land in different quarters and sizes.

One-offs and comparability breaks

Two disclosed items distort a naive year-over-year read of the statements, and both must be normalized before extrapolating.

Sources: FY2023 charge and EPS impact [28]; FY2024–FY2025 [29]; the FY2025 six-month program and its $308M FY2026 Q1 tail [30] [31]. FY2027E assumes no new program.

The realignment is disclosed in the FY2025 Form 10-K, which now defines the three reportable operating segments as the Americas, EMEA and Asia Pacific and continues to split revenue by consulting versus managed services [32]. One further, favorable normalization: FY2023 adjusted EPS also excluded a $0.38 investment gain, so the underlying operating comparison is cleaner than the headline GAAP EPS suggests [33].

The forward base case: FY2026–FY2027

FY2026 is effectively de-risked: three of four quarters are reported, with nine-month revenue of $55.5 billion, up 7% in dollars and 4% in local currency [34]. Management's June 18, 2026 guidance narrowed the full year to 3%–4% local-currency growth (4%–5% excluding an estimated 1-point drag from the U.S. federal business), GAAP operating margin of 15.3% (+60 bps), adjusted operating margin of 15.8% (+20 bps), GAAP diluted EPS of $13.38–$13.50 and adjusted EPS of $13.78–$13.90, with free cash flow of $10.8–$11.5 billion and at least $9.5 billion returned to shareholders [35] [36]. My FY2026E simply sits at the midpoints.

FY2027 has no company guidance, so the base case below is built driver-by-driver. It lands close to — and modestly below the initial trajectory of — external consensus, which currently centers near $76.6 billion of revenue and $14.67 of adjusted EPS; those out-year estimates have been drifting lower, not higher, over the past quarter.

Sources: FY2026E figures are management guidance [37] [38]; FY2027E is a modeled base case extending the disclosed drivers, with consensus used only as a reasonableness check (as compiled).

How the FY2027 numbers are derived. Start from the FY2026 base of ~$73.5 billion. Hold local-currency growth at ~4% — the middle of the FY2026 run-rate, on the judgment that discretionary consulting keeps stabilizing and advanced-AI work keeps scaling, while the U.S. federal drag normalizes rather than deepens — and assume neutral FX, giving ~$76.5 billion. Extend the adjusted margin cadence one more notch to ~16.0%. Assume the six-month optimization program is finished, so the GAAP line converges up toward the adjusted line (no severance wedge). That yields adjusted operating income near $12.2 billion; after ~24.5% tax and a ~1% lower share count from ongoing buybacks, adjusted EPS lands around $14.65, roughly 6% growth. A reader who disagrees — say, who expects only 2% growth or no further margin gain — can re-run the same arithmetic and will get a materially lower EPS; that sensitivity is the point.

Sources: reported FY2023–FY2025 GAAP and adjusted EPS [39] [40]; FY2026E per guidance midpoints [41]; FY2027E modeled. GAAP and adjusted converge as one-offs roll off.

Reconciliation to guidance. My FY2026E ($13.84 adjusted, $13.44 GAAP, ~$73.5B revenue, 15.8% adjusted margin) sits at the midpoints of the June 2026 guidance and needs no departure. Worth noting the trajectory of guidance itself: the initial FY2026 outlook a year earlier (Sept 2025) was for 2%–5% local-currency growth and adjusted EPS of $13.52–$13.90 — the range has since been narrowed and the low end trimmed up, so management has firmed the year even as the top of the EPS range held [42]. My only departure from the Street is on FY2027: I model ~$14.65 adjusted EPS versus consensus near $14.67 — essentially in line — but I flag that consensus revisions have been downward, so I would not assume upside to it.

Cash conversion and capital return

Cash is the least ambiguous part of the story. Free cash flow swung from $8.6 billion in FY2024 (on softer operating cash flow of $9.1 billion) to a record $10.9 billion in FY2025, comfortably exceeding net income — conversion above 1.4x [43]. Guidance calls for $10.8–$11.5 billion again in FY2026 and at least $9.5 billion of capital returned via dividends and buybacks, which is what steadily shrinks the share count and adds a point or so to EPS growth each year [44].

Sources: FY2023–FY2025 free cash flow and dividends/buybacks as reported [45]; FY2026E free cash flow and capital-return floor per guidance [46]; FY2027E modeled.

Segment view

Accenture's reportable segments are its three geographies; because of the FY2025 realignment, a clean cross-year geographic margin comparison is not available, so the segment read below is the FY2025 snapshot on the new basis, alongside the consistent consulting-versus-managed-services split. In FY2025 the Americas (half of revenue) ran a ~15% GAAP operating margin (16% adjusted), EMEA ~13%, and Asia Pacific the most profitable at ~18% [47].

Source: FY2025 segment revenue, operating income and margins, Q4 FY2025 results [48]. Only FY2025 is shown on the new basis: after the realignment, prior-year "North America / EMEA / Growth Markets" figures are not directly comparable and Accenture did not restate full-year FY2024 geography [49].

The forward-looking segment signal to watch is type of work, not geography: in the first nine months of FY2026 managed services grew 6% in local currency against consulting's 2%, continuing the multi-year shift toward the more annuity-like, slower-decaying revenue base [50]. That mix shift supports the margin-and-cash durability the forward estimate leans on, even if it caps the pace of any top-line reacceleration.

What would make the estimate wrong

The base case is not fragile on margins or cash — those have three years of steady, disclosed progression behind them and are guided for FY2026. It is most exposed on the top line:

Bookings are the leading indicator, and they are soft. New bookings fell 1% in FY2025 and 2% again in Q3 FY2026, with book-to-bill hovering around 1.2. If bookings do not re-accelerate, the ~4% FY2027 revenue assumption is the first thing to cut [51].

The U.S. federal drag could widen. Management already carves out an estimated 1-point hit to FY2026 growth from federal; a deeper or more prolonged pullback would push growth below the guided range [52].

AI is a two-sided risk. The same advanced-AI demand driving the FY2025 reacceleration also threatens to deflate the labor-hours Accenture bills for; the base case assumes AI remains net-additive to bookings, which is the assumption most likely to be tested [53].

On balance, the recent trajectory is durable enough to support the forward estimates on margin, EPS and cash, where the record is steadiest — but the top line is guided to decelerate, and a base case that assumes stabilization rather than reacceleration is the honest one to carry.