A massive AWS outage in October 2025 affected more than 3,500 companies in over 60 countries and made it clear how much the global operations depend on cloud infrastructure.
In this article, we will cover the very recent internet outage statistics, the most affected companies, historical outages, the economic impacts, and the resilience strategies defining the future of connectivity.
Internet Outage Statistics (2026): Key Highlights
- AWS Outage on Oct 20, 2025, affected 3,500+ companies.
- Over 6.3 million outage reports were from the US alone for the AWS Outage.
- Snapchat logged 3 million AWS outage reports.
- Global network outages rose 33.38% from January to May 2025.
- IT downtime costs averaged $14,056 per minute in 2024.
- The US faces a $458 million loss during a one-hour internet outage.
- Power failures caused 45% of total outages in 2025.
- 86% organisations adopted multi-cloud strategies for resilience.
- Mean Time To Restore (MTTR) averaged 80 minutes globally.
- AWS, Azure, and Google Cloud controlled 63% of the infrastructure market.
Trending: Read our Cloudflare statistics to uncover more about its outage history and impact.
Internet Down Today: The Latest Major Outage Incident
- On October 20, 2025, AWS experienced a major outage affecting its US-EAST-1 region, disrupting many downstream services worldwide.
- The outage began around 3 am ET and lasted for many hours; AWS reported all services returned to normal by 18:01 ET (≈22:01 UTC).
- The root cause was traced to a DNS resolution failure in the US-EAST-1 region, related to an empty DNS record and incorrect automation updates.
- Multiple major websites and apps (including gaming and streaming services) reported issues during this period.
- A single DNS failure paralysed 3,500+ companies across 60+ countries on October 20, 2025.
- Over 4 million outage reports were submitted within just 2 hours of the outage commencement.
- Peak outage reported 6.3 million reports from the US alone and 1.5 million from the UK.
The table below shows outages reported by country:
| Rank | Country | Outage Reports |
|---|---|---|
| 1 | United States | 6.3M |
| 2 | Great Britain | 1.5M |
| 3 | Germany | 774K |
| 4 | Netherlands | 737K |
| 5 | Brazil | 589K |
| 6 | France | 587K |
| 7 | Australia | 516K |
| 8 | Canada | 475K |
| 9 | India | 428K |
| 10 | Japan | 368K |
- Downdetector recorded a 970% increase in average daily baseline reports during the AWS outage peak.
- The recovery was phased and inconsistent across different regions and services, lasting throughout the day.
- With 5.65 billion people using the internet (as seen in our internet usage report), even short-lived cloud outages can translate into millions of disrupted users and significant economic fallout
AWS News Today: Cloud Outage Updates & Affected Companies
- Snapchat received approximately 3 million outage reports, the single most-affected platform during the October 2025 AWS outage.
AWS itself experienced 2.5 million user reports on Downdetector, reflecting the direct impact on the cloud platform.
WhatsApp, DoorDash, Disney+, the McDonald’s App, and many other major brands were also victims of the outage.
Design platforms such as Canva, which serve about 220 million monthly active users, show how outages hit both creators and small businesses that rely on cloud-based tools.
| Rank | Service | Outage Reports |
|---|---|---|
| 1 | Snapchat | 3M |
| 2 | AWS | 2.5M |
| 3 | Roblox | 716K |
| 4 | Amazon | 698K |
| 5 | 397K | |
| 6 | Ring | 357K |
| 7 | Instructure | 265K |
| 8 | Venmo | 154K |
| 9 | Fortnite | 142K |
| 10 | Canva | 133K |
Source: Ookla
Is Amazon Down? Recent Amazon Outage And Service Disruptions
- Amazon.com’s retail platform reported 698k+ user reports during the October 2025 outage.
- AWS Lambda serverless computing services were completely unavailable during the failure.
- DynamoDB NoSQL database service triggered the initial cascade failure affecting 1000+ downstream services.
- The outage’s reach is amplified by scale: AWS, Amazon’s fastest-growing arm that earned $30.9 billion in Q2 2025, underpins thousands of downstream services, so regional failures ripple widely
Source: TOI
Major Outages Over Time: Cloud Outage Statistics & Historical Data
- Global network outages increased 33.38% from January to May 2025, rising from 1382 to 1843 incidents.
- March saw a 32% surge in global outages, reaching 2110 total incidents.
- The US-centric outages peaked at 55% of global outages in early 2025, significantly above the historical 40% baseline.
- June 2025 documented 1219 global outages, signalling stabilisation after peak deployment periods.
The table below shows month-wise global outages:
| Month | Global Outages |
|---|---|
| January | 1,382 |
| February | 1,595 |
| March | 2,110 |
| April | 1,804 |
| May | 1,843 |
| June | 1,219 |
- AWS maintained market leadership with 30% global cloud infrastructure in Q2 2025.
- Comparatively, Microsoft Azure held 20%, and Google Cloud reached 13% market share.
- The “Big Three” cloud providers collectively controlled 63% of the global enterprise cloud infrastructure market in Q2 2025.
- August 2024 IAM authentication issue caused regional login failure lasting 50 minutes.
- On June 2, IBM cloud users experienced a widespread 7-hour and 48-minute-long login outage, affecting console, CLI, and API access, as well as cloud resource management and support portal functionality.
- Ingram Micro suffered a ransomware attack from July 3-10, 2025, causing a nearly weeklong outage that halted order processing and shipping operations.
- Google Cloud and Cloudflare had a Service Control overload and dependency issue on June 12, 2025, triggering multi-hour outages affecting apps like Spotify, Discord, and Cloudflare services.
Source: CRN 2, Cherry Servers, CRN 1, Thousandeyes
Economic & Business Impact Of Cloud Downtime
- IT downtime costs averaged $14,056/ minute in 2024.
- Large enterprises face $23,750/ minute in downtime costs, higher than standard averages.
- High-priority applications cost $67,651/hour of downtime for enterprises, whereas normal-priority applications cost $61,642/hour in unplanned downtime.
- The United States would face $458,941,744 in losses during a single hour of internet outage.
- The United Kingdom would incur $136,297,189 during a 1-hour internet disruption.
- A 24-hour internet outage in the US would cost her $11,014,601,859.
The table below shows the losses different nations would incur if they had internet outages:
| Country | 1-Hour Outage (USD M) | 5-Hour Outage (USD M) | 10-Hour Outage (USD M) | 24-Hour Outage (USD M) |
|---|---|---|---|---|
| United States | 458.94 | 2,294.71 | 4,589.42 | 11,014.60 |
| China | 412.38 | 2,061.89 | 4,123.78 | 9,897.06 |
| United Kingdom | 136.3 | 681.49 | 1,362.97 | 3,271.13 |
| Japan | 113.1 | 565.52 | 1,131.03 | 2,714.48 |
| Germany | 62.46 | 312.28 | 624.56 | 1,498.94 |
| India | 59.63 | 298.13 | 596.27 | 1,431.04 |
| South Korea | 52.95 | 264.76 | 529.51 | 1,270.83 |
| France | 36.3 | 181.52 | 363.03 | 871.28 |
| Italy | 26.8 | 134.01 | 268.02 | 643.25 |
| Canada | 24.03 | 120.13 | 240.25 | 576.61 |
- Hourly downtime costs exceed $300,000 for 90% firms.
- 41% of the enterprises say hourly downtime costs them $1 million to over $5 million.
- 12.5% of the companies (1 in 8) lose over $10 million monthly from outages and service degradations.
- Outage frequency increased 12% from 43% to 51% of companies experiencing $1 million+ monthly losses YoY.
- Compounding the risk, cyber threats remain relentless, with over 2,200 attacks occurring daily worldwide, which raises the stakes for robust incident detection and recovery strategies.
Source: Help Net Security, Calyptix, Cloudzero, Veeam, Big Panda
Causes Behind Rising Internet Outages
- DNS resolution failure affecting DynamoDB API endpoints triggered the October 2025 AWS cascade.
- DNS failover issue in monitoring tools, combined with Network Load Balancer health-check failures.
- A race condition in the DNS management system occurred between the DNS Planner and DNS Enactor components.
- An invalid automated update to Google’s API management system caused the June 2025 Google Cloud outage.
- 154 Big Tech data centre permits filed in Virginia in the first nine months of 2025 alone.
- Power issue remains the most common cause of outage, with 45% of the incidents in 2025.
Cooling issues have caused 14% of the total outages.
Here are the different causes of outages in 2025:
| Cause of Outage | Percentage |
|---|---|
| Power | 45% |
| Cooling | 14% |
| Network | 11% |
| IT systems (hardware/software) | 9% |
| Fire suppression | 6% |
| Information security-related | 5% |
| Unknown | 4% |
| Fire | 2% |
| Third-party provider | 2% |
| Colocation provider | 1% |
Source: Uptime Intelligence, Business Insider
Strengthening Outage Resilience
- 86% organisations report implementing multi-cloud strategies for resilience.
- Multi-cloud reduces single-point-of-failure risk and minimises unplanned downtime.
- An outage in one cloud won’t impact services deployed on alternative providers.
- Cloud computing workloads can be routed to alternate providers when the primary fails.
- Multi-region deployments eliminate single points of failure through a distributed architecture.
- Mean Time To Restore (MTTR) averages 80 minutes across industry benchmarks.
- 50% of data centres experienced at least one impactful outage over the past three years.
- Organisations achieving 99.9% availability experience a maximum of 9 hours of downtime annually.
- AWS’s average incident resolution time is 1.5 hours for 38 incidents (Aug 2024-Aug 2025).
- Google Cloud incidents average 5.8 hours despite reporting 78 incidents during the same period.
- Azure incidents average 14.6 hours, though the company reported fewer total incidents.
Source: Cherry Servers, Flexera, Data Foundry, Uptime Intelligence, Launchdarkly, VexxHost
