Adobe beat its own Q3 FY2026 guidance with revenue up 12.9% to $6.76 billion and non-GAAP EPS up 15.4% to $6.13, but roughly 2 points of growth came from the Semrush acquisition, a 6.8% smaller share count supplied about half the EPS gain, and the forward gauges — 8.4% RPO growth and an unchanged 10.2% full-year ARR target — all point to deceleration.
Revenue beat the high end of guidance, but the recurring-revenue target behind it did not move
Adobe reported third-quarter fiscal 2026 results on Sept. 10, 2026, for the quarter ended Aug. 28, 2026. (Adobe's fiscal year ends in late November, so this "Q3 2026" covers June through August 2026.) Revenue of $6.760 billion rose 12.9% from $5.988 billion a year earlier and came in $40 million above the top of the $6.67–6.72 billion range management had guided in June. Non-GAAP earnings per share of $6.13 beat the guided $6.05–6.10 by three cents.
Two things complicate that headline. About two points of the 12.9% growth is bought rather than earned: Adobe closed its $1.874 billion acquisition of Semrush on April 28, 2026, which added roughly $480 million of annual recurring revenue. And management raised the full-year revenue and EPS targets while leaving the full-year target for ending annualized recurring revenue growth — the forward run-rate of the subscription base — unchanged at 10.2%. The raise reflects the quarter that happened, not a better view of the quarters ahead.
Operating margin is the share of revenue left after paying the costs of running the business, before interest and tax. Adobe's "non-GAAP" version strips out stock-based compensation, acquisition costs and the amortization of acquired intangible assets; the GAAP version does not.
Takeaway: Adobe beat its own quarter and raised the year, yet every forward-looking gauge in the release either stood still or slowed — the FY2026 recurring-revenue growth target held at 10.2%, contracted-but-unrecognized revenue (RPO) grew only 8.4% against revenue's 12.9%, and Q4 revenue guidance implies growth decelerating to about 10%. Strip out Semrush (~2 points) and the share-count reduction (~7 points of the 15.4% non-GAAP EPS growth), and the underlying business is growing at roughly 11% with margins going the wrong way.
Where the growth came from
Adobe now reports along two customer groups rather than the old Digital Media / Digital Experience segments — a disclosure change it flagged in December 2025 when it said FY2026 reporting would center on customer-group subscription revenue and total company ending ARR.
Business Professionals & Consumers (Acrobat, Adobe Express, Acrobat Studio) grew subscription revenue 16% to $1.91 billion, 15% excluding currency moves. This is the smaller group at 29% of customer-group subscription revenue, and it has grown faster than the other in every quarter Adobe has disclosed the split (+15% in Q3 and Q4 FY2025, +16% in Q2 and Q3 FY2026).
Creative & Marketing Professionals (Photoshop, Illustrator, Premiere, plus the Experience Platform/Experience Manager marketing stack) grew 13% to $4.65 billion, 12% excluding currency. This is where Semrush landed: Adobe's 10-Q describes the April 2026 acquisition as enhancing Adobe Experience Manager with search-engine and generative-engine optimization.
Semrush's contribution can be sized from Adobe's own disclosure: in Q2 it added about $40 million of customer-group subscription revenue in roughly one month of ownership, which annualizes to about $120 million a quarter — close to 2 points of the 12.9% total growth. A further point came from currency: Adobe reported 13% growth "or 12% in constant currency," meaning a weaker dollar inflated the reported figure. That tailwind is narrowing — in Q2 the gap was two points (13% reported vs. 11% constant currency). Net of both, organic growth at constant currency is running around 10%.
Total ARR exiting the quarter was $27.50 billion, up from $27.10 billion at the end of Q2. ARR — annualized recurring revenue, the yearly value of subscription contracts in force — is Adobe's headline demand metric, and Adobe restates the prior year at current exchange rates so ARR growth is effectively currency-neutral. At Q2, that growth was 12.5%, including the ~$480 million from Semrush; without the acquisition it was closer to 10.5%.
The cost of running AI is showing up in the margin
The clearest deterioration in this quarter is on the cost side, and it is visible at two levels.
Gross margin fell 0.6 points because the cost of delivering subscriptions grew 24.1% (to $633 million from $510 million) against subscription revenue growth of 13.7%. Adobe has not yet filed the Q3 10-Q — the release says it expects to do so in September 2026 — so there is no Q3 breakdown of that line yet. In the Q2 10-Q, Adobe attributed essentially all of the same line's increase to "hosting services and data center costs" (16 of the 16 percentage points of growth), and defines that line as including "AI inferencing costs," i.e. the compute bill for running generative models on customers' behalf. Compensation added 5 points and lower amortization subtracted 6.
Operating margin fell further, 1.5 points on a GAAP basis to 34.8%. Research and development rose 18.4% to $1.288 billion (19.1% of revenue, up from 18.2%) and general and administrative rose 19.6% to $488 million — the non-GAAP reconciliation shows $18 million of acquisition-related expense in the quarter against none a year ago, which accounts for part of the increase but not most of it. Sales and marketing was the one restrained line, up 11.5% — slower than revenue.
Worth noting which direction the adjustments cut: non-GAAP operating margin fell more than GAAP margin (−2.3 points vs. −1.5 points). That is because the items Adobe excludes shrank as a share of revenue — mainly stock compensation, which grew only 4.4% to $544 million, well below revenue. In other words the GAAP margin decline understates the operating cost pressure; the measure management prefers is the one that looks worse.
Net income was flat; EPS was not
GAAP net income rose just 3.1% while GAAP EPS rose 10.5%. Two separate effects account for the gap between those numbers and between them and the 7.8% growth in pre-tax income.
Tax. The effective tax rate was 22.5% this quarter against 19.0% a year ago. The prior-year quarter carried a $97 million discrete tax benefit (Adobe excluded it from non-GAAP results as a $0.24 per-share "income tax adjustment"); this year's equivalent was $2 million. Held at last year's rate, net income would have been about $1.91 billion — up 7.8%, in line with pre-tax income — and GAAP EPS about $4.83. So the apparent profit stall is mostly a comparison artifact, not a deterioration.
Share count. Diluted shares fell 6.8% to 395 million. Adobe repurchased about 9.5 million shares for $2.232 billion of cash in the quarter — 92% of the $2.44 billion of free cash flow it generated (operating cash flow of $2.523 billion less $85 million of capital spending). Because Adobe pays no dividend, buybacks are the entire capital return. Roughly 7 of the 15.4 percentage points of non-GAAP EPS growth came from the smaller denominator rather than from higher profit; non-GAAP net income itself grew 7.6%.
Non-GAAP EPS is the cleaner year-over-year comparison here because Adobe applies a fixed 18% tax rate to it in both periods, which removes the tax distortion above — but it does not remove the buyback effect.
One incidental data point on the buyback: $2.232 billion spent on approximately 9.5 million shares implies an average of roughly $235 a share, against roughly $257 implied in the year-ago quarter ($2.057 billion for about 8.0 million shares). Adobe is retiring more stock for less money because the stock is cheaper, which mechanically raises the per-share benefit of a given dollar of buyback.
What the forward indicators say
RPO grew 8.4%, to $22.16 billion from $20.44 billion. Remaining performance obligations are revenue under contract that has not been recognized yet — a rough proxy for the order book. A year ago Adobe highlighted RPO "accelerating to 13 percent year-over-year growth"; this year it grew at roughly two-thirds of that pace, and below the 12.9% growth in recognized revenue. RPO also fell in absolute terms from $22.52 billion at the November 2025 fiscal year-end, and enterprise renewal timing makes the line lumpy quarter to quarter, so one reading is not a verdict — but 8% forward growth against 13% reported growth is the single most important tension in this release. The current portion held at 67% of the total, unchanged from a year ago.
The FY2026 ARR growth target was left at 10.2%, exactly where June's release put it, even as revenue and EPS targets went up. Applied to the $25.20 billion of ARR Adobe exited FY2025 with, 10.2% implies about $27.77 billion at year-end — only about $270 million above the $27.50 billion just reported, versus the $400 million added in Q3 itself. The guidance embeds a slower fourth quarter for new recurring revenue, not a faster one.
AI monetization is growing fast off a small base, and the disclosure got thinner. Adobe led the release with "AI-first ARR grew more than 150% year over year" and a milestone of one billion monthly active users across its creativity and productivity products, but gave no dollar figure. In June it did: AI-first ARR "triples year over year and exceeds $500 million" — under 2% of total ARR at the time. Dropping the dollar figure while keeping the growth rate is a choice worth watching; a >150% growth rate is also a step down from June's roughly 200%.
Guidance
Target
Q4 FY2026
vs. Q4 FY2025 actual
Total revenue
$6.80B – $6.85B
+9.8% to +10.6% (from $6.19B)
Business Professionals & Consumers subscription revenue
For the full year, Adobe now targets revenue of $26.576–26.626 billion (from $26.50–26.60 billion in June), GAAP EPS of $18.12–18.17 (from $17.90–18.00), non-GAAP EPS of $24.45–24.50 (from $24.35–24.45), ending ARR growth of 10.2% (unchanged) and non-GAAP operating margin of about 45.0% (unchanged). Against FY2025's $23.769 billion of revenue and $20.94 of non-GAAP EPS, that is 11.8–12.0% revenue growth and about 17% non-GAAP EPS growth.
Two details in that raise matter more than its size. First, the $51 million increase at the midpoint of the full-year revenue target is smaller than the Q3 beat that produced it — the implied Q4 number actually slipped by roughly $15 million at the midpoint from what June's guidance embedded. Nothing about the fourth quarter got better. Second, the guided FY2026 non-GAAP operating margin of ~45.0% is about 1.2 points below the 46.2% Adobe delivered in FY2025 ($10.99 billion of non-GAAP operating income on $23.769 billion of revenue), so margin compression is a full-year condition, not a one-quarter blip.
The GAAP-to-non-GAAP gap for the year is $6.33 per share, of which $5.42 is stock-based compensation and about $0.41 is one-off items: a $70 million non-cash goodwill impairment on the Publishing & Advertising reporting unit ($0.18) and legal loss contingencies ($0.23), both recognized earlier in FY2026.
Leadership is turning over at the same time as growth
Three senior changes landed within three months of each other, all confirmed in Adobe's own SEC filings:
CEO succession. On Sept. 2, 2026 — eight days before these results — the board appointed Anil Chakravarthy, currently President of the Customer Experience Orchestration business, as President and CEO effective Dec. 1, 2026, the first day of fiscal 2027. Shantanu Narayen, currently Chair and CEO, retires as CEO and becomes Executive Chair of the board.
The CFO seat is interim. Dan Durn left on June 15, 2026; Steve Day, a 20-year Adobe finance veteran, has been interim CFO since, and it is his name on the guidance in this release.
A divisional president is leaving. Also on Sept. 2, David Wadhwani notified Adobe he is stepping down as President of the Creativity & Productivity business effective Sept. 27, 2026, staying on as a senior advisor.
Adobe also disclosed on July 14, 2026 that it had granted General Counsel Louise Pentland a retention agreement with severance protections that sunset 12 months after the new CEO starts — the standard mechanism for holding an executive team together through a transition, and a sign the board was managing retention risk before the succession was public.
Our read
The operating business is doing what a mature subscription software company does when its growth rate is under pressure: acquiring (Semrush, at roughly 3.9 times the ~$480 million of ARR it brought), leaning on the faster and cheaper end of the product line (Acrobat and Express at +16%), and buying back stock aggressively enough to convert single-digit profit growth into mid-teens per-share growth. None of that is unsound, and the cash generation supporting it is real: $2.52 billion of operating cash flow in a single quarter on a 34.8% GAAP operating margin.
What we would not extrapolate is the 13% top line. It contains about 2 points of acquisition, about 1 point of currency, and it sits above both an 8.4% order book and a 10.2% recurring-revenue target that management declined to raise. The plausible trajectory from here is revenue growth converging down toward the ARR growth rate — roughly 10% — with margins settling one to two points lower than FY2025 as AI inference costs scale with usage rather than with price. The open question the next two quarters should answer is whether AI-first products, growing over 150% but from well under 2% of ARR, get large enough to change that arithmetic before the acquisition and currency contributions lap out. A new CEO arriving on Dec. 1 with an interim CFO beside him will be setting FY2027 targets against exactly that question.
Source: Adobe Inc. Form 8-K filed Sept. 10, 2026 (Exhibit 99.1, Q3 FY2026 earnings release, quarter ended Aug. 28, 2026); Form 10-Q for the quarter ended May 29, 2026 (filed June 15, 2026); Form 8-K filed Sept. 8, 2026 (leadership changes); Form 8-K filed Dec. 10, 2025 (Exhibit 99.1, Q4/FY2025 results). The Q3 FY2026 Form 10-Q had not been filed as of Sept. 13, 2026; Adobe stated it expects to file in September 2026, and figures in that filing could differ from the release.