Adobe Paid $150 Million for Deliberate Friction, Exposing a Crack in Subscription Models
On March 13, 2026, Adobe announced two significant developments simultaneously: a $150 million settlement with the U.S. Department of Justice and the Federal Trade Commission, and the retirement plans of CEO Shantanu Narayen after 18 years at the helm. The coincidence of these announcements was not accidental in terms of narrative but doesn't alter the underlying analysis: a company with over $19 billion in annual revenue built a system over years where canceling a subscription was deliberately difficult. That decision, which likely retained millions in short-term revenue, ultimately cost much more than just money.
The central accusation from the DOJ and the FTC was specific: Adobe offered annual plans billed monthly but buried penalties for early cancellation — equivalent to 50% of remaining payments — in fine print, nested hyperlinks, or hover icons. The cancellation process involved multiple screens, call transfers, and, in some cases, continued charges even after the customer believed they had canceled. Adobe denies any wrongdoing and claims to have improved its processes. The settlement includes $75 million in cash to the DOJ and another $75 million in free services for affected customers, pending court approval.
What interests me is not the litigation but the financial logic that made it possible.
Retained Revenue from Friction Comes at a Deferred Cost
When 95% of your revenue comes from subscriptions — as has been the case with Adobe for several years — every percentage point of retention has a direct impact on cash flow. The math is straightforward: if you have a million active subscribers with an average monthly ticket of $55, an additional point of monthly retention equates to keeping 10,000 subscribers who would have otherwise left. That’s $550,000 monthly or $6.6 million annually for each point. Scaled to Adobe’s actual base, the numbers are substantially larger.
From that perspective, designing frictions into the cancellation process appears to be a financially rational short-term decision. The customer who cannot exit easily continues to generate revenue. The problem is that this revenue is not validated by the market; it is being retained by the system's design. That difference has consequences that do not show up on the income statement until years later.
The first consequence is regulatory, and it has already materialized: $150 million plus legal costs accumulated since the original lawsuit in June 2024. The second is reputational, and its impact on future customer acquisition is harder to quantify but equally real. The third — and most relevant to any CFO — is the distortion of internal metrics. A company that retains customers through friction does not know precisely how many of those customers would continue paying if the exit process were simple. That uncertainty contaminates any projection of recurring revenue.
When the Model's Architecture Depends on Difficult Exits
Adobe's model is not unusual. The structure of "annual plan billed monthly" is standard in the subscription software industry precisely because it combines two advantages: the perception of a low monthly price for the customer and an annual financial commitment for the company. The model works well when customers perceive enough value to renew voluntarily. It works poorly — and creates regulatory risk — when retention depends on customers not finding the way out.
Adobe generated over $19 billion in revenue in 2024, with 95% coming from subscriptions. The $150 million settlement represents approximately 0.75% of that annual revenue: a material hit but not an existential one in accounting terms. However, the market processed it differently: Adobe's shares fell by 5.62% on the announcement day. This drop does not reflect the $150 million. It reflects the risk that simplifying cancellation — now an implied requirement of the settlement — will reduce the customer retention rate of those who were not staying by conviction but by inertia.
Here lies the most crucial data for reading this case correctly. If Adobe needed frictions to maintain its retention metrics, then a portion of its recurring revenue was not being sustained by the perceived value of the product but by the cost of exit. That distinction separates a financially robust model from one that silently accumulates risk in its customer base.
The competitive context amplifies the problem. Adobe faces pressure from alternatives like Canva, AI-based image generation tools, and suites like Affinity, which directly compete on price and offer more flexible payment models. In that environment, a customer trapped in an Adobe plan due to friction and not value is precisely the customer most likely to migrate to a competitor as soon as they can exit cleanly.
The Real Cost is Not in the Court Settlement
The $150 million figure tends to dominate the coverage because it's concrete and large. But the most significant financial cost to Adobe in the coming quarters does not come from the check to the DOJ. It arises from two effects that will activate simultaneously.
First, the company must redesign its cancellation flows to make them more straightforward and transparent. That’s operationally manageable but introduces a variable that internal financial models likely did not calibrate accurately: how many existing customers will cancel when the process is simple. If that rate is materially higher than the models projected, the impact on recurring revenue for the next fiscal year will be visible in the quarterly reports, irrespective of the court settlement.
Second, the $75 million in free services for affected customers creates a situation that requires careful management. If those services include extended access to premium products, there is a risk that customers who receive them will develop usage habits they won’t maintain when the free period ends. In the best-case scenario, those customers convert. In the worst case, Adobe absorbs the delivery cost without subsequent retention.
The leadership transition announced on the same day adds a layer of operational complexity. The subscription model that Narayen built over 18 years — taking revenues from $4 billion in 2013 to over $19 billion in 2024 — now requires an adjustment that his successor must execute under regulatory scrutiny and with retention metrics more exposed than ever.
The subscription software industry is observing this settlement closely. The FTC has already applied similar pressure to other companies, and the "click-to-cancel" rule established in 2024 turns cancellation ease into a regulatory standard, not an optional design decision. Any company that has built its retention rate on difficult exits is operating with a liability that is yet to be recognized on its balance sheet.
Retention That Is Not Earned Becomes Debt
Adobe's settlement is not an anomaly. It is the predictable result of optimizing retention from the wrong side of the equation. When a company designs the exit process to be more costly than staying, it is not building loyalty; it is accumulating dissatisfied customers who remain due to inertia or ignorance of their options. That customer base does not renew with conviction, does not refer other users, and does not absorb price increases without additional friction.
The only retention that has sustainable financial value is that which occurs because the customer actively chose to stay, having had a real option to leave. A customer who pays $55 monthly knowing they can cancel in two clicks is worth more, in terms of projected lifetime value, than one who pays the same amount because they didn't find the cancellation button. The first is an asset. The second is a disguised liability, and when that disguise comes off — due to regulation, competition, or both — the impact arrives directly on cash flow without warning.
