Invest1 publisher3 min readPublished
Adobe's $27.50bn of ARR works out to $27.50 a year for each of its billion monthly users
Adobe reported record third-quarter revenue of $6.76bn and its billionth monthly active user on September 10. By Shantanu Narayen's own split, freemium expansion explains only about half of the lowered ARR outlook.
The Investor · Invest desk

What happened
- Adobe reported record third-quarter revenue of $6.76bn on September 10, up 13% year over year and 12% in constant currency, for the quarter that ended August 28, 2026.
- Monthly active users across Acrobat, Creative Cloud, Express and Firefly passed one billion for the first time, up from roughly 850 million when Adobe reported first-quarter results in March.
- Total ARR exiting the quarter was $27.50bn, and remaining performance obligations reached $22.16bn, with 67% of that figure classified as current.
- Adobe's shares declined in after-hours trading after the record report.
- Anil Chakravarthy becomes president and CEO on December 1, 2026, succeeding Shantanu Narayen, who announced in March that he was retiring from the top role.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
- contradiction Narayen's own split puts only about half the reduced ARR outlook on freemium expansion, and the other half on Creative Cloud price increases Adobe chose to defer, so treating the free tiers as the whole cause follows half the company's accounting.
- decision The faster of Adobe's two levers is reinstating those deferred price increases. That cost falls on the customers who already pay, not on the free cohort that drove the user milestone.
- cost Compute for free generations is paid in the quarter it is consumed while any subscription revenue from those users arrives later, so the cost lands in current operating margin.
- constraint ARR is close to next year's revenue, so a shortfall in the ARR line now caps fiscal 2027 growth however large the monthly user count gets.
Adobe exited the quarter with $27.50bn of annual recurring revenue and one billion monthly active users [9][5]. Divide the first by the second and each monthly active carries $27.50 of ARR a year, about $2.29 a month [1][2]. Roughly 150 million of those users arrived since March, when the count stood near 850 million [5][3].
The payoff case, as techtimes.com frames it, is that a 2-to-5% conversion rate on a billion users implies a pool of 20 million or more paying subscribers [15]. I would price the narrower number instead. Creative freemium monthly actives crossed 90 million in the second quarter and grew more than 70% year over year [14], which against the new total is about 9% of the billion [4].
"Reaching a landmark of more than one billion monthly active users is a defining moment for Adobe," Narayen said in the earnings release [12]. Adobe's own account of the ARR drag has two halves: on the second-quarter call, Narayen attributed the reduced ARR outlook roughly half to deferred Creative Cloud price increases and half to accelerating freemium expansion [17]. A deferred price increase is a decision Adobe can reverse in a quarter. A free Firefly user converts on their own schedule, and under the generative credits system that user spends one credit per standard image and significantly more on a video generation before converting at all [16]. Steve Day, the interim CFO, said of the third quarter: "We are expanding our user base through a freemium strategy and deepening engagement with agentic experiences to deliver long-term durable growth." [18]
Whatever the free tiers cost to serve, they sit inside a non-GAAP operating margin of 43.9% and a GAAP margin of 34.8%, on operating income of $2.97bn and $2.35bn against $6.76bn of revenue [7][1][5][6]. Operating cash flow was $2.52bn against $2.20bn a year earlier, up 14.5% [8][7]. techtimes.com puts no figure on the infrastructure bill for serving hundreds of millions of free monthly users, though it describes that expense as real and near-term [21].
Four times $6.76bn is $27.04bn, so ARR of $27.50bn sits 1.7% above the current revenue run rate [8]. techtimes.com attributes the after-hours decline to investors pricing how long it takes to convert a free Firefly user into a paying subscriber while Adobe absorbs the GPU cost of every generation [19]. Non-GAAP diluted EPS was $6.13 against $5.31 a year earlier, and the shares fell anyway [6].
If the deferred Creative Cloud price increases land in fiscal 2027, about half the drag reverses without a single free user converting, on Narayen's own split [17]. If the 90 million creative freemium cohort keeps compounding above 70% and its conversion shows up in the ARR line, the suppression was prepayment [14]. What would prove me wrong is duller than either: ARR growth reaccelerating while the free tiers stay exactly as wide as they are now, which would mean the cost to convert was never the binding constraint. In the same quarter Adobe repurchased roughly 9.5 million shares [11].
What to watch
- Whether the deferred Creative Cloud price increases appear in fiscal 2027 guidance, which on Narayen's split accounts for about half the reduced ARR outlook.
- The next disclosure of the creative freemium cohort, and whether Adobe starts reporting a conversion rate out of it.
- The first ARR outlook Chakravarthy owns after taking over on December 1, and whether the free tiers stay as wide as they are.