Adobe: The Subscription and the Software

By Turnaround Radar

In April 2023, a Reddit user posted a screenshot of their Adobe cancellation flow. Seven screens. Three upsells. A $150 early termination fee buried behind a hyperlink inside a tooltip. The post got 47,000 upvotes. Three years later, Adobe paid $150 million to settle the DOJ case that screenshot helped inspire.

Here is the part that matters: the software that user was trying to cancel still has a 4.5-star rating on every app store it touches. Photoshop Express holds 4.53 stars across 2.4 million Google Play ratings. Lightroom sits at 4.5. Creative Cloud aggregate reviews on Capterra score 4.6 out of 5. The products are not the problem.

Adobe's Trustpilot rating is 1.2 out of 5.

That is not a typo. Seven thousand reviews, and 83% of them are one star. Not about the software. About the subscription. The cancellation. The charge they did not expect. The chat agent that would not stop offering discounts instead of processing the refund.

Two Adobes exist. The stock is down 38% because the market cannot decide which one is dying.

All current verdicts: The Verdict Board →

See the Investment Council's verdict on $ADBE →

The short answer is fear. Specifically, AI fear.

Adobe peaked at $421 in mid-2025, riding the same generative AI wave that lifted every software company with a chatbot. Then the narrative flipped. If Midjourney could generate a marketing banner in four seconds, why pay $55/month for Photoshop? If Canva's 260 million monthly users could produce professional designs with zero training, where was Adobe's moat? If Figma — the company Adobe tried to buy for $20 billion before regulators killed the deal, forcing Adobe to write a $1 billion termination check — was now public and eating the UI/UX market, what exactly was the growth premium paying for?

The fear compounded in three waves. First: Shantanu Narayen, CEO for 18 years — the architect of the cloud transition that took Adobe from boxed software to $26 billion in recurring revenue — announced his retirement on March 12, 2026. No successor named. Just a board committee and a vague promise that David Wadhwani, head of Digital Media, was the internal favorite. The stock dropped 7% that day, despite Adobe simultaneously reporting record Q1 revenue.

Second: Mizuho's Gregg Moskowitz downgraded Adobe to Neutral on April 27, cutting his target to $270 from $315. His note was blunt: "We wrongly held off from downgrading." He cited intensifying competition in the prosumer and SMB segments, margin erosion risk, and no clear catalyst. The stock fell another 5%.

Third: the broader AI rotation. Capital migrated from "AI beneficiaries" to "AI builders" — from Adobe and Salesforce to Nvidia and Palantir. Adobe's forward PE collapsed from a five-year average of 32.6x to 10.6x. At 10.6 times forward earnings, Adobe trades cheaper than Procter & Gamble. A company growing revenue 12% with 90% gross margins is priced like a utility.

Michael Burry bought in the low $250s. His Q1 2026 13F filing showed new positions in Adobe, MercadoLibre, PayPal, and Lululemon. He called Adobe a "fat pitch" — citing Firefly's commercial-safe AI, deep enterprise integration, and what he described as "credit-driven selling exhaustion."

The question is whether Burry is right, or whether the 1.2 Trustpilot rating is the more honest signal.

FY2026 Guidance: Revenue $25.9–26.1B, Non-GAAP EPS $23.30–23.50.

Q2 FY2026 Guide: Revenue $6.43–6.48B, Non-GAAP EPS $5.80–5.85.

These are not the numbers of a company in distress. Revenue grew 12%. Margins expanded. Cash flow hit a record. The $25 billion buyback announced April 21 — replacing a similarly sized program that was nearly exhausted — represents roughly a quarter of Adobe's $103 billion market cap. At the current share price, that retires approximately 96 million shares, or 23% of the float, over four years.

The central question of this report is whether Adobe's consumer reputation problem is about the software or the subscription. We test this directly.

Null hypothesis (H₀): The proportion of 1-star reviews is the same on billing/service platforms (Trustpilot, PissedConsumer) and product-quality platforms (Google Play, Capterra).

Alternative hypothesis (H₁): The 1-star proportion is higher on billing/service platforms.

Data:

Result: Z = 421.66, p < 10⁻²³

The difference is 78.4 percentage points (95% CI: 78.1%–78.8%). On platforms where users rate the product — the editing tools, the filters, the workflow — Adobe scores 4.5 stars and above. On platforms where users rate the experience of paying for it, canceling it, or disputing a charge, Adobe scores 1.2.

This is not a product crisis. It is a billing-relationship crisis. The $150 million DOJ settlement, the FTC complaint, the 2,000+ BBB complaints in three years — all of it traces to subscription friction, not software quality.

The bear case rests on the assumption that AI competition will eventually erode Adobe's revenue growth. We test whether the deceleration is already statistically visible in the quarterly ARR data.

Data: Digital Media ARR year-over-year growth rate, Q1 FY2024 through Q1 FY2026 (9 quarters).

Mann-Kendall trend test: S = −27, Z = −2.71, p < 0.05. Sen's slope = −0.16 percentage points per quarter.

The trend is real but slow. At the current pace, Digital Media ARR growth declines by roughly 0.6 percentage points per year — from 12.8% to 11.5% over two years, projecting to ~10.9% by Q1 FY2027. This is deceleration, not collapse. Adobe is not losing subscribers. It is adding them at a marginally slower rate each quarter.

The market, however, has priced in collapse. A forward PE of 10.6x implies either growth going to zero or margins compressing by 20+ points. Neither is happening.

The numbers miss three things.

First, the billing reputation is a ticking clock. The $150 million DOJ settlement requires Adobe to reform its cancellation flow, clearly disclose early termination fees, and notify users before converting free trials to paid subscriptions. These are material changes to a business model that generates $26 billion in recurring revenue. If the reforms reduce involuntary churn (users who stay because canceling is too hard), the short-term ARR impact could be negative even though the long-term customer relationship improves. This is a temporary revenue headwind disguised as a governance win.

Second, the CEO transition is not priced correctly. David Wadhwani, the internal frontrunner, built Adobe's Digital Media business — the division responsible for 74% of revenue. His appointment would signal continuity, not disruption. But the market is pricing the transition as if Narayen is irreplaceable. The 7% earnings-day drop on March 12 — in the face of record revenue — was entirely about the retirement announcement. When the successor is named, the uncertainty premium should partially unwind.

Third, Firefly's monetization is earlier than the market realizes. AI-first ARR more than tripled year-over-year in Q1 FY2026. Firefly has generated 24 billion assets and is adopted by 75% of the Fortune 500. The generative credit model ($0.05–$0.30 per generation) creates a usage-based revenue layer on top of the subscription base. Firefly generated $400 million in direct revenue between 2024 and 2025. That is still less than 2% of total ARR, but the growth trajectory — tripling annually — suggests the contribution becomes material by FY2027.

Recovering:

NOT recovering:

Unknown:

Trustpilot selection bias. Trustpilot and PissedConsumer are complaint-driven platforms. Users with negative billing experiences are disproportionately likely to leave reviews. The 1.2 rating reflects the worst sliver of Adobe's customer base, not the median experience. However, the volume — 7,068 reviews on Trustpilot alone — makes it a statistically meaningful signal, not anecdotal noise.

ARR growth data limitations. The 9-quarter series has limited statistical power. The Mann-Kendall test detects a trend at the 5% level, but with only 9 data points, a single strong quarter could flip the signal. The Q1 FY2025 rebound (12.6% after three quarters of decline) already demonstrates volatility within the trend.

Canva comparison asymmetry. Canva's 260M MAU and $4B ARR are real. But Canva and Adobe serve fundamentally different user segments. Canva dominates the prosumer and SMB market — the 90% of use cases that don't require professional-grade output. Adobe dominates the 10% that does. The overlap is smaller than the market assumes.

Forward PE compression may be structural. If the market has permanently re-rated software companies from 30x to 15x forward earnings, Adobe at 10.6x may not be as cheap as the historical comparison suggests. The 5-year average of 32.6x includes the zero-rate, growth-at-any-price era that is unlikely to return.

Bear case (30% probability): The DOJ cancellation reforms accelerate churn. Wadhwani's appointment underwhelms. Q2 earnings miss on the top line as Canva converts SMB accounts. Firefly monetization stalls at <3% of ARR. The stock tests $200 — a 20x compression from peak.

Base case (50% probability): Q2 earnings meet guidance. Wadhwani is named CEO by Q3. Buyback absorbs selling pressure. Firefly ARR reaches $1B by FY2027. The stock grinds to $300–340 over 12 months — roughly in line with the analyst consensus target of $327.

Bull case (20% probability): Q2 earnings beat with Firefly acceleration. New CEO named quickly with a credible AI-first vision. DOJ reforms are implemented smoothly. Burry's "fat pitch" thesis plays out. Multiple re-rates from 10.6x to 18–20x. Stock recovers to $380–420.

Now ($262): This is a high-quality franchise trading at a once-in-a-decade valuation. But the catalysts are binary and imminent. The risk/reward favors patience — not conviction.

June 11 (Q2 earnings): This is the decider. If Adobe beats on revenue and guides Firefly ARR above $600M annualized, the re-rating begins. If the company misses or cuts FY2026 guidance, the "AI kills Adobe" narrative hardens and $220 is in play.

H2 2026 (CEO announcement): The second catalyst. If Wadhwani gets the job and articulates an AI-first platform strategy (not just Firefly as a feature, but Adobe as the AI creative operating system), the uncertainty discount fades.

Position sizing matters more than timing. The expected value is modestly positive (+11%), but the distribution is wide. This is a half-position now, full position after June 11 clarity.

When Adobe reports Q2 on June 11, subscribers will get the Turnaround Radar earnings breakdown within 24 hours. Three numbers matter:

1. Firefly ARR disclosure. If Adobe breaks out Firefly/AI-first ARR and it exceeds $500M annualized, the "AI threat" narrative flips to "AI beneficiary." Watch for this specific line in the earnings release.

2. Net new ARR. Adobe guided $2.6B in net new ARR for FY2026 — the highest target ever. Q1 delivered approximately $700M. If Q2 tracks at $650M+ despite the DOJ cancellation reforms taking effect, the churn fear is overblown.

3. CEO successor timing. The earnings call Q&A will be the first opportunity for the board to signal timeline. Any mention of "weeks, not months" changes the calculus.

Subscribe to make sure you don't miss the June 11 analysis.

All current verdicts: The Verdict Board →

See the Investment Council's verdict on $ADBE →