By Industry
Streaming service outage cost: Netflix, Spotify, Disney+ benchmarks
Streaming outages cost between $80,000 and $1.2 million per hour depending on service and time of day. The math is simple in theory (ARPU times affected concurrent users times hours) and difficult in practice because concurrency varies by an order of magnitude between off-peak and prime-time, and ad-supported services have a different cost curve from subscription services. The most expensive moment to be down is the kickoff of a major live event.
Per-Service Benchmarks
Estimated per-hour outage cost by streaming service
These figures triangulate from publicly disclosed ARPU and subscriber counts, peak-hour concurrent-user estimates, and ad-revenue-per-stream data. They are estimates, not disclosed numbers. The peak-prime figures assume a major outage at peak viewing time (8 to 11 PM local in the largest markets, or kickoff of a major live event). The lower-bound figures assume off-peak hours.
| Service | Subscribers | ARPU | $/hr (off-peak) | $/hr (peak) | Model |
|---|---|---|---|---|---|
| Netflix (peak prime) | 260M global | $11.99 avg | $800,000 | $1,200,000 | Subscription |
| Disney+ | 150M global | $8.50 avg | $250,000 | $450,000 | Subscription + Hulu bundle |
| Spotify Premium | 236M premium | $5.40 avg | $150,000 | $300,000 | Subscription |
| Spotify (ad-supported) | 390M free tier | Ad-served minutes | $100,000 | $250,000 | Ad revenue per stream |
| YouTube (ad-served) | 2.5B MAU | Variable | $1,000,000 | $4,000,000 | Ad impressions per minute |
| Twitch (live) | 31M DAU | Sub revenue + ads | $80,000 | $200,000 | Streamer subs + ads |
| HBO Max / Max | 100M global | $10.50 avg | $200,000 | $350,000 | Subscription + ad-supported tier |
ARPU and subscriber data from each company's most recent quarterly reporting (Netflix Q4 2025, Disney Q4 2025, Spotify Q4 2025, etc.). Per-hour figures are our triangulation, not disclosed. Estimated, as of 2026-05-18.
Disney Plus Launch, 12 November 2019
The reference event for a major streaming launch outage
Disney Plus launched on 12 November 2019 with widespread login and playback failures lasting most of launch day. Disney had pre-sold over 10 million subscribers before launch, and the actual day-one signup spike exceeded their internal forecasts. The infrastructure ramp could not match real demand, and the result was a launch-day outage that was widely covered in the press.
Disney did not disclose a direct dollar cost. Three signals are useful. First, the share price briefly dipped 1.4% on the news, then recovered within two trading days, suggesting markets viewed the outage as a non-material short-term setback rather than a strategic concern. Second, the launch nonetheless added approximately 26.5 million subscribers by the end of Q1 2020 (Disney's disclosure), so the outage did not visibly slow the medium-term subscriber ramp. Third, the press coverage was largely sympathetic ("they had too much demand"), so the brand-damage line was small.
The cost-shape lesson: launch-day outages of highly-anticipated consumer services tend to cost less than the per-hour models predict because the public narrative is "they were overwhelmed by demand", which is reputationally neutral or even positive. Outages mid-service-life, where the narrative is "they failed at something they should be doing well", cost meaningfully more per hour.
Live-Event Outage Multiplier
The Super Bowl, the World Cup, and the live multiplier
The most expensive moment to be down as a streaming service is the kickoff of a major live event. Three reasons compound. First, concurrency peaks at 5 to 10x the average daily peak, so the revenue-density-per-minute is much higher. Second, the cost of missing live content is qualitatively different from missing on-demand content. A viewer who can't watch Stranger Things now will watch it tomorrow. A viewer who can't watch the Super Bowl now has missed the cultural moment, and rebooking is not possible. Third, social-media attention to streaming outages peaks during live events, so the reputation cost spikes.
Notable live-event streaming failures include Hulu's 2019 Super Bowl outage (10 minutes of buffering during the second half), Peacock's 2024 NFL Wild Card Sunday Night Football (intermittent issues), and Amazon Prime Video's 2022 NFL Thursday Night debut (load issues in the first quarter). None resulted in disclosed dollar figures, but all produced significant press coverage and follow-on subscriber retention questions.
For a streaming service that has paid for premium live content rights (typically $300M to $2B per multi-year deal), an outage during the live event partially defeats the strategic purpose of the rights deal. The economic loss is therefore measured against the marginal value of the deal, not just against the per-hour ARPU math. For a $1B deal with 100 live events over the contract, a single major-event outage can be modelled as $10M+ of strategic value loss in addition to the direct revenue line.
Ad-Revenue Mechanics
Why Spotify and YouTube outages cost differently
Subscription streaming services lose deferrable revenue when they are down (the monthly fee accrues regardless). Ad-supported streaming services lose non-deferrable revenue (an ad impression that did not happen at 8 PM cannot be made up). The cost-per-hour math is therefore different. For Spotify's premium tier, an outage at 8 PM technically loses nothing in subscription revenue, although it accelerates churn risk. For Spotify's ad-supported tier, an outage at 8 PM loses the ad revenue that would have served against the streams that did not happen.
YouTube is the largest example. Roughly $7B per quarter in ad revenue (per Alphabet's Q4 2025 disclosure) amortises to approximately $3.2M per hour averaged across all hours. Peak-hour ad revenue is meaningfully higher, perhaps $5M to $8M per hour. A major YouTube outage at peak viewing time costs $4M to $8M per hour in lost ad revenue alone, before the brand-damage and creator-relationship lines.
For a streaming product manager, the practical implication is that ad-supported tiers warrant tighter monitoring SLOs than subscription tiers because the revenue loss is irrecoverable. The investment case for additional reliability spend is stronger for the ad-supported product line.
Twitch and the Live-Streaming Creator Economy
A second-order cost layer: the creator economy
Twitch (and to a lesser extent YouTube Live) has a second-order cost layer that traditional streaming services do not have. When the platform is down, individual creators lose subscription revenue, donation income, and the in-the-moment audience attention they have invested marketing dollars to acquire. Creators whose income depends on the platform routinely report multi-thousand-dollar single-event losses from major Twitch outages.
Twitch's direct revenue loss during an outage is small (DAU times ad and sub revenue). The creator economy loss is potentially much larger and feeds back to Twitch as platform-departure risk. After major outages, top creators publicly evaluate alternative platforms (YouTube Live, Kick), which is a brand-damage cost that does not appear in standard outage cost models. For platform owners managing creator-economy services, the right cost model includes a creator-departure-risk premium on top of the direct revenue loss.
Frequently Asked
Common Questions
How much does a Netflix outage cost?
How much did the Disney Plus launch-day outage cost?
How much does a Spotify outage cost?
Why are live-event outages more expensive than on-demand outages?
What was the cost of the PSN outage in 2011?
How is streaming outage cost different from generic SaaS outage cost?
Related