Nobody knows the exact amount, and the estimates change every few months

The internet does not have a single counter that tracks how much data exists on it. Every estimate you see — whether it says 64 zettabytes or 120 zettabytes or some other number — is a calculation based on guesses about how much data different companies and services store, how many devices are connected, and how much new data gets created each day. Different researchers using different methods get different answers, and all of them are partly wrong.

The reason the estimates keep growing is not just that people are uploading more. It is also that we are getting better at measuring what was already there, that companies are storing more copies of the same data in different places, and that new types of devices (security cameras, sensors in factories, phones in people's pockets) are constantly generating data that did not exist before. A single video uploaded to YouTube might be stored in dozens of data centers around the world at once, so counting it once versus counting each copy separately changes the total dramatically.

Key Takeaways

  • Internet data is measured in units like terabytes, petabytes, and zettabytes, where each step up is 1,000 times larger than the one before.
  • The total amount of data on the internet is estimated somewhere between 60 and 175 zettabytes, depending on what you count and who is doing the counting.
  • Most of that data is not text or images you see — it is video files, backups, logs of what happened on servers, and copies stored in multiple locations.
  • The amount of new data created each day is growing faster than storage space, which is why companies delete old data and why internet services sometimes lose your information.

How data size is measured

Data is measured in bytes. One byte is enough to store a single letter or number. From there, the units jump by factors of 1,000 each time you go up: a kilobyte is 1,000 bytes, a megabyte is 1,000 kilobytes, a gigabyte is 1,000 megabytes, and so on.

By the time you reach the scale of the entire internet, you are talking about zettabytes. One zettabyte is 1,000 exabytes. One exabyte is 1,000 petabytes. One petabyte is 1,000 terabytes. A single terabyte is enough to store roughly 250 movies in high definition, or about 500 hours of video. So one zettabyte is 250 million times that.

To put a zettabyte in perspective: if you downloaded one gigabyte of data per second without stopping, it would take you roughly 32 million years to read one zettabyte. The entire internet is estimated to contain somewhere between 60 and 175 zettabytes, depending on what you count.

What counts as "internet data"

The confusion about how much data exists comes partly from disagreement about what should be counted. Does it include only data that is publicly accessible, or also private databases that companies keep behind firewalls? Does it count each copy of a file separately, or just once? Does it include data that is stored but never accessed, like old backups?

Most estimates count data that is stored somewhere connected to the internet, whether or not the public can see it. That includes email servers, cloud storage, social media databases, video streaming libraries, and the logs that servers keep of every request they receive. A single large company like Google or Amazon stores petabytes of data — Google alone is estimated to store somewhere between 10 and 15 exabytes, though that number is not official.

Video makes up the largest share of internet data by far. YouTube, Netflix, TikTok, and similar services store enormous libraries of video files, and each file is often stored in multiple formats and multiple locations. A single two-hour movie might exist as a 4K version, a 1080p version, a 720p version, and a mobile version, each stored in several data centers on different continents. That one movie could account for gigabytes of storage space across the internet.

Why estimates are unreliable

Internet data estimates come from a few sources: companies like IDC and Statista that survey businesses and extrapolate from samples, academic researchers who model data growth, and occasionally statements from large tech companies about their own storage. None of these methods captures the whole picture.

A company that stores data might not know exactly how much it has, especially if data is spread across many servers and data centers. A researcher estimating total internet data has to guess about how much data is stored by companies that do not publish those numbers. And the definition of "the internet" itself is fuzzy — does it include data on private corporate networks that happen to use internet protocols? Does it include data on devices that are not always connected?

The estimates also change because storage technology is improving faster than data is growing. A company might compress its video library or delete old backups, shrinking the total amount of data stored even though new data is being created. Or a company might start storing more redundant copies for safety, growing the total without creating any new information.

How much new data is created each day

Estimates suggest that roughly 2.5 quintillion bytes of data are created every single day — that is 2.5 exabytes, or about 2,500 petabytes. That includes everything: emails sent, videos watched, photos taken, messages posted, sensor readings from industrial equipment, financial transactions, and logs of server activity.

Most of that data is not something a person created intentionally. When you watch a video on YouTube, YouTube logs which video you watched, when you watched it, how long you watched, what device you used, and where you were located. When you search Google, Google logs your search term, your location, and your account information. When you use a credit card, the bank logs the transaction, the merchant logs it, and payment processors log it. Each of those logs is data that gets stored somewhere.

The rate of new data creation is growing faster than the rate at which storage capacity is growing. That means companies have to make choices about what to keep and what to delete. Most services keep recent data and delete old data after a certain period. Email providers might keep your messages for years, but your login logs might be deleted after 90 days. Video services keep popular videos but might delete videos that nobody watches.

Where internet data is actually stored

Internet data is not stored in one place. It is spread across thousands of data centers owned by companies like Amazon Web Services, Microsoft Azure, Google Cloud, and many others, as well as data centers run by individual companies for their own use. A data center is a building full of servers — computers that store data and respond to requests from other computers.

Large companies store the same data in multiple data centers in different geographic locations. If one data center catches fire or loses power, the data still exists in other places. This redundancy means the same file might be counted multiple times in estimates of total internet data. A photo you upload to Google Photos might be stored in data centers in the United States, Europe, and Asia simultaneously.

Data centers use enormous amounts of electricity to power the servers and to cool them down — servers generate heat, and heat damages equipment. The energy cost of storing and serving data is one reason why companies delete old data and why internet services sometimes lose your information. Keeping data costs money, so companies keep only what they think they will need.

Why the estimates matter less than you think

The exact total amount of data on the internet does not affect how the internet works or how fast your connection is. What matters more is how much data is being created and stored in specific places, and how that affects the companies that run those services.

When estimates say the internet contains 100 zettabytes or 150 zettabytes, they are usually trying to make a point about how much data is being generated and how important data storage and processing have become to the economy. The number itself is less important than the trend: data creation is accelerating, storage is becoming more centralized in the hands of a few large companies, and the energy cost of storing and serving that data is growing.

For a person using the internet, what matters is whether the service you use has enough storage for your data, whether it backs up your data in case something goes wrong, and whether it deletes your data when you ask it to. Those are questions about specific companies and services, not about the internet as a whole.

Frequently Asked Questions

How much data does one person create per day?

A typical person using email, social media, and streaming services generates somewhere between 1 and 5 gigabytes of data per day, though most of that is not data you intentionally created — it is logs and metadata that services record about your activity. A person who takes photos, records videos, and uses cloud storage might generate more.

Is all the data on the internet actually useful?

No. Much of it is redundant copies, old backups, logs that nobody will ever read, and spam. Companies keep data because it is cheaper to store it than to decide what to delete, and because they might need it for legal reasons or to recover from failures. A significant portion of internet data is probably never accessed after it is created.

How fast is internet data growing?

Estimates vary, but most researchers say the total amount of data on the internet is growing by 20 to 30 percent per year. That is faster than the growth in storage capacity, which means companies have to get better at compression and deletion, or spend more money on storage.

Do I own the data I upload to the internet?

Legally, you own the copyright to content you create, but the service you upload it to owns a license to store and display it. Read the terms of service for the specific platform — they vary widely. Some services promise to delete your data when you ask, others keep it longer, and some keep it indefinitely even after you delete your account.

Why do companies need to store so much data?

Companies store data for several reasons: to serve content to users, to train artificial intelligence systems, to analyze user behavior for advertising, to comply with legal requirements, and to recover from failures. Deleting data costs money in terms of deciding what to keep, so many companies keep everything unless there is a specific reason to delete it.