The Subscription Fatigue Report: What Users Really Pay For

The narrative about subscription fatigue usually goes: people are drowning in subscriptions, they are cutting them, SaaS is in trouble. The survey data I collected across Q4 2025 and Q1 2026 tells a more nuanced story that I think is more useful for both users and the companies building subscription products.

The Survey

1,200 knowledge workers, recruited from LinkedIn communities, tech forums, and social media, completed a survey on their software subscription spending. Demographics: 68% work at companies of 50-500 employees, 32% are freelancers or solo operators. Geography: 71% US, 15% EU, 14% rest of world. All respondents have at least five active software subscriptions.

The survey covered: total monthly spend on software subscriptions, number of active subscriptions, satisfaction ratings per category, subscriptions cancelled in the past 12 months, and the stated reasons for cancellations.

The Spending Reality

Average monthly software subscription spend per respondent: $312. This includes both personal and work-related subscriptions where the respondent is paying personally.

Median: $178. The mean is significantly higher than the median because a subset of respondents (approximately 15%) spend over $500/month, pulling the average up.

The category breakdown by average spend:

• Productivity and work tools: $94/month
• AI tools: $67/month (the fastest-growing category)
• Creative and design tools: $48/month
• Security and privacy tools: $39/month
• Learning and education: $31/month
• Other: $33/month

The AI tools category barely existed in equivalent surveys from 2022. Its emergence as the second-largest spending category in three years is a remarkable adoption story.

The Cancellation Data

1,047 of 1,200 respondents (87%) had cancelled at least one subscription in the past 12 months. The average number of cancellations was 3.2 per respondent.

The stated reasons for cancellation are where the data gets interesting:

• "I stopped using it" - 41%
• "I found something better" - 23%
• "Price went up beyond what felt fair" - 19%
• "Too many subscriptions, trimming" - 11%
• "Company went in a direction I did not like" - 6%

The dominant reason for cancellation is not price - it is cessation of use. Users cancel subscriptions they have stopped using, not subscriptions that are too expensive. This runs counter to the "subscription fatigue is about cost" narrative.

The implication for SaaS companies: retention is fundamentally a usage problem, not a pricing problem. Users who are actively using and getting value from a product are remarkably price-tolerant. Users who have stopped using a product are cancelling regardless of price.

What Users Will Not Cut

When asked which subscriptions they would be last to cancel, the patterns were clear.

Communication and collaboration tools (Slack, Zoom, Microsoft Teams, Google Workspace) scored highest in retention intent. The network effect lock-in is real - you cannot stop using a tool that everyone you work with also uses.

AI tools scored second highest. ChatGPT Plus was the most frequently cited specific subscription that respondents would not cut. The second most cited was Claude Pro. The pattern: users who had found genuine daily use cases for AI tools were highly retentive. Users who had not found use cases but subscribed out of curiosity or FOMO were high-churn.

Password managers. 1Password and Bitwarden Premium were cited as among the hardest to cancel - once users have stored credentials and important data in a password manager, the switching cost is high enough to ensure strong retention.

The design tools users chose to invest in. For power users of design tools, the specialization and workflow integration created strong retention even at higher prices.

The "Too Many Subscriptions" Myth

Only 11% of cancellations were attributed to general subscription fatigue (too many subscriptions). The vast majority of cancellations were specific: a specific product stopped being used, or a specific product was replaced by a better one.

This is important because "subscription fatigue" is often used as a catch-all explanation for SaaS churn that obscures more actionable diagnoses. If users are leaving because they stopped using the product, the solutions are about activation, onboarding, and habit formation - not pricing or packaging.

What the AI Category Data Shows

The AI tools category deserves specific attention because the retention pattern is distinctly bimodal.

Among respondents who use AI tools more than 3 times per week: retention intent is extremely high. 91% said they would not cancel their primary AI subscription even if the price doubled.

Among respondents who use AI tools less than once per week: 67% had cancelled at least one AI subscription in the past 12 months, with "stopped using it" as the dominant reason.

The implication: AI tools have an onboarding problem more than a pricing problem. Users who discover regular use cases for AI tools become highly retentive customers. Users who sign up and never develop the habit churn quickly. The retention battle is won or lost in the first 30 days of the subscription.

Practical Takeaways

For individual users auditing subscriptions: The useful question is not "what am I spending?" but "what am I using and what am I getting from it?" Run a personal audit against the last 30 days of actual usage, not what you intended to use. Cancel the unused ones without guilt - the cancellation reason category most likely to improve your subscription relationship is "I stopped using it," and it has an obvious fix.

For SaaS companies: The activation problem is bigger than the pricing problem for most products. If users are cancelling because they stopped using the product, the question to answer is: why did they stop using it? What use case did they think they were buying that they failed to achieve?

Comparing subscription costs across categories is easier than ever with the feature breakdowns available at best AI tools and best project management tools. The right subscription is not the cheapest one - it is the one you will use consistently enough to justify the cost.