Dark Web
Deep Web vs Dark Web: What Is the Difference?
The deep web and the dark web are constantly confused — and the confusion is not harmless, because it makes an ordinary part of the internet sound sinister and a small hidden part sound enormous. They are different things, for different reasons, and the difference is easy once you see it.
The one-sentence difference
The deep web is everything search engines do not index. The dark web is a small part of the deep web whose sites are deliberately hidden and need special software to reach. Everything on the dark web is part of the deep web, but almost everything in the deep web is not part of the dark web — it is your email, your bank, and the article behind a paywall.
Think of it as a library. The surface web is the books on the open shelves, listed in the catalogue. The deep web is the archive rooms: real content, mostly boring, kept behind a staff door because it is private or restricted. The dark web is a locked room inside the archive that you can only enter with a special key — and the people inside chose it that way.
What lives in the deep web
You use the deep web every day, probably without noticing. Any page that a search engine cannot reach is deep web, and the reasons are mostly mundane:
- Your private accounts — email, online banking, shopping orders, medical portals, cloud storage. A search engine cannot log in as you, so it cannot index these pages.
- Paywalled and subscription content — newspaper articles, academic papers, streaming catalogues, and databases that only subscribers can read.
- Internal systems — a company intranet, a university portal, a government employee system. They are on the internet, but not public.
- Unlinked pages — a page that no other page links to is effectively invisible to a search crawler, even if it is not password-protected.
- Generated pages — search results, booking confirmations, and database queries that only exist for a moment, in response to your request.
None of this is hidden in the secretive sense. It is unindexed because indexing it would be a privacy violation or a technical impossibility. Your bank wants search engines kept out of your account. That is the deep web doing its job.
What lives on the dark web
Dark web sites are different in kind, not just in degree. They are built on networks such as Tor, use .onion addresses, and are reachable only with software such as Tor Browser. The hiding is the point: the site operator does not want the server location known, and the visitor does not want their location known either.
Documented uses include anonymous tip lines run by news organisations, whistleblowing platforms, privacy-focused email and chat services, libraries and archives, and forums for people living under censorship. Illegal marketplaces exist too, and researchers and law enforcement report them — but they are one part of a much smaller ecosystem than the headlines imply.
Side by side
| Question | Deep web | Dark web |
|---|---|---|
| Why is it not in search results? | It needs a login, or nothing links to it, or it is generated on demand | It is deliberately hidden on a separate network |
| How do you reach it? | Any browser, usually with an account | Special software such as Tor Browser |
| Is it illegal? | No — almost all of it is ordinary and legal | The technology is legal in most countries; what you do there decides |
| How big is it? | Large — most of your online life sits behind logins | Small — thousands of onion services, not millions |
| Typical example | Your email inbox | An anonymous tip line run as an onion service |
The 90 percent myth
The most repeated number about the deep web is that it makes up 90 to 95 percent of the internet, with the surface web as the tiny tip of an iceberg. It appears on vendor blogs, in news articles, and in explainers from companies that should know better. It is wrong — or more precisely, it was never right in the form it is quoted.
The figure traces back to a 2001 estimate by researcher Michael Bergman, made when the web was far smaller and far less understood. It was an estimate of database content, not a measurement of the internet, and it has been repeated for more than two decades without anyone re-measuring it — because nobody can. So much of the deep web sits behind logins and private systems that no one has a reliable way to count it. Security researchers have explicitly debunked the figure, noting that it has no current basis, and the Fortinet documentation cited by other vendors puts the dark web itself at less than 0.01 percent of the internet.
How to judge any size claim. Ask three questions: who counted it, how did they count it, and when? If a page quotes a percentage of the internet with no method and no date, it is repeating folklore, not reporting a fact. A serious claim names its source and its method.
The honest answer about size is that the deep web is big and the dark web is small — and that any page claiming a precise percentage of either is guessing. Our companion guide explains in full why the counts disagree instead of quoting one number as fact.
Why the confusion persists
Three forces keep the two terms tangled. First, the iceberg image is vivid and easy to draw, so it survives even when the numbers behind it are debunked. Second, vendors and news sites repeat each other: a figure quoted on one site becomes a source for the next, and after enough repetitions it looks authoritative. Third, "dark web" sounds dramatic, so headlines use it for anything unindexed, including perfectly ordinary deep web content.
The practical effect is that ordinary people overestimate the dark web and underestimate the deep web. They imagine a vast criminal underworld, when the reality is a large body of private everyday content plus a small hidden network with mixed uses. Getting the terms right is not pedantry — it is the difference between fear and understanding.
What this means for you
- Your deep web is normal. Email, banking, health portals and subscriptions are deep web. There is nothing to worry about and nothing to "access" — you already live there.
- The dark web needs a deliberate step. You cannot wander into it. Reaching it means downloading Tor Browser from torproject.org and choosing to visit an onion address.
- Legality follows behaviour, not layers. Reading your own email is legal; buying illegal goods is illegal, on any layer of the internet.
- Size claims deserve scepticism. Any page that gives you a precise percentage of the internet without a method and a date is repeating a myth.
Sources
- Tor Project — A short introduction to Tor (Tor Specifications, accessed 14 November 2026)
- Tor Metrics — Servers: relays and bridges (accessed 14 November 2026)
- Onion services — Wikipedia (accessed 14 November 2026)
The 2001 Bergman estimate and its debunk are discussed in the research report behind this site, which cites CloudSEK's correction of the 90–95 percent figure. Last checked: 14 November 2026.
Frequently asked questions
What is the deep web?
The deep web is everything on the internet that search engines do not index: your email inbox, online banking, paywalled articles, private documents, a company intranet. Most of it is ordinary, everyday content that needs a login.
What is the dark web?
The dark web is a small part of the deep web. Its sites are deliberately hidden on networks such as Tor, use .onion addresses, and need special software such as Tor Browser to reach. Unlike most of the deep web, they are hidden by design, not just unindexed.
Is the deep web 90 percent of the internet?
That figure traces back to a 2001 estimate that was never a measurement, and researchers have been correcting it for years. Nobody can measure the deep web reliably, because so much of it sits behind logins. Honest sources refuse to quote a percentage as fact.
Is the deep web illegal?
No. Almost everything in the deep web is legal and mundane: your email, your bank, your medical portal, a subscription newspaper. Illegality depends on content and behaviour, not on which layer of the internet something sits in.
Can search engines find deep web content?
Only the parts that are public and linkable. A search engine cannot log in to your email or your bank, cannot read a paywalled article it has no access to, and cannot reach a page that no other page links to. That is why the deep web exists: it is unindexed, not hidden.