Media & Publishing Churn Rate: Benchmarks & Analysis

The news industry's retention data shows a strong correlation between newsletter engagement and subscription retention. Subscribers who open a daily or weekly newsletter at least 3 times per week retain at significantly higher rates, making email the primary retention channel rather than app visits. The New York Times credits its Cooking and Games verticals as major retention drivers - users with cross-product engagement (news + cooking, or news + Wordle) churn at half the rate of news-only subscribers.

Introductory pricing creates a structural churn event when promotions expire. Publications that use hard cut-offs (e.g., $1/month for 3 months, then full price) see sharp cancellation spikes at the expiration date. Those that use a gradual step-up pricing model - slowly increasing from promotional to full price over 6 months - reduce this churn event by 30-40%. Proactive outreach 30 days before price increases, paired with value reminders, also measurably reduces price-shock cancellations.

Beyond the top two drivers, the next three reasons in the data are low engagement - not reading consistently (20%); news fatigue and avoidance behavior (10%); competing subscription (6%), each meaningful enough to deserve its own retention initiative when an operator's monthly cancellation feedback shows that pattern concentrating in a single cohort. Consumer-app retention curves bend most sharply at the day-7 and day-30 marks, so cohort analysis that stops at month-1 misses the long-tail engagement decay that drives most of the eventual cancellation, particularly in subscription-heavy categories where annual plans defer the cancellation event without reducing the underlying disengagement. The most useful next step for any operator above their category benchmark is reading the cancellation feedback verbatim rather than aggregating it into reasons, because the language users actually choose at the cancel screen reveals the trust event sooner than the categorized counts ever will.