do a website's terms bind a scraper?

this is general information, not legal advice. most of this post is about EU law. the names lawyers use for online terms, like clickwrap and browsewrap, come from US court cases, so a short section covers those too.

TL;DR: it depends on how you met the terms

almost every site has terms of use, and many of them say "no scraping". whether that line binds you depends less on what it says and more on one question: did you ever agree to it? here are the questions we hear most, each with a one-line answer. the rest of the post explains them.

• I never clicked "I agree". do the site's terms bind me? → often not. but no EU rule settles it: it depends on the country, and on whether you knew the terms.
• I opened an account on the site. do its terms bind me? → in most cases, yes. ticking a box or clicking "sign up" next to the terms is how online contracts are made.
• does a "no scraping" line in the footer make scraping illegal? → no. it can matter in a contract dispute, and it tells you what the site wants. it isn't a law.
• what is the worst case if I break terms I did accept? → a contract claim: the site can ask a court to make you stop and pay damages. it isn't a crime by itself.
• the site wrote to me and quoted its terms. does that change anything? → yes. now you know the terms. US courts have held scrapers to terms after a letter like that.
• can I scrape through my own logged-in account? → only if that account's terms allow automated access.

how terms get accepted online

a contract needs two sides that agree. on a website, nobody signs anything, so courts look at what the site asked you to do and what you did. over the years, lawyers have given each pattern a name. you will meet these names in contracts, court rulings and your DPO's questions, so here they are, from the clearest yes to the weakest, plus one old one.

• clickwrap. you tick a box or click "I agree" next to the terms before you can go on. it is the clearest way to show you agreed.
• scrollwrap. like clickwrap, but you must scroll through the whole text before the "I agree" button works.
• sign-in wrap. a line under the sign-up button says "by creating an account, you agree to our terms", with a link. there is no box to tick: signing up is the yes. some courts call it a hybrid of clickwrap and browsewrap.
• browsewrap. the terms sit behind a link, usually in the footer, and they say that using the site means you accept them. nobody asks you anything.
• shrinkwrap. the old one: terms printed inside a software box, accepted when you keep and use the product. you will see the word in older rulings; it rarely comes up for websites.

the names are only labels. a court looks at the real page: was the notice easy to see, and did you do something that clearly meant yes?

what EU law says

those names come from US courts, and we get to them below. the EU has no single law on website terms. whether a contract was made is mostly a question for each country's own contract law. but a few EU rules and rulings set the frame, and the good news is that they are easy to follow.

a click can make a contract. in 2015 the Court of Justice of the European Union (CJEU), the EU's top court, ruled that clicking to accept terms that open in a new window can count as agreeing "in writing" to a clause in them, as long as you could print and save the terms before you clicked. the case was about which court gets a dispute, but it shows how EU courts see a click. in 2022 the same court went further for the same kind of clause: a written contract that links to the terms is enough, even with no box to tick.

a site can limit the use of its data by contract. in the Ryanair case (2015), the airline's site made visitors tick a box to accept terms that banned scraping for commercial use. the CJEU ruled that terms like these can work even when the data has no copyright or database right to protect it. whether a contract was really made is still up to each country's own law.

sites have duties too. when a contract is made online, its terms must be made available in a way you can save and print. for consumers, a term they had no real chance to read before the contract can be unfair, and an order button must say clearly that you will pay. most of these rules protect consumers. a business that scrapes gets less of that protection.

one country's view: 🇩🇪 Germany. German law has a strict rule for bringing standard terms into a contract: the other side must be told clearly and get a fair chance to read them (§ 305 BGB). that rule protects consumers; between businesses it doesn't apply, and ordinary contract rules decide. and in 2014 Germany's Federal Court of Justice (BGH), its top court for civil cases, found that a flight portal did not compete unfairly just by scraping an airline's site against its terms, even though booking on that site meant ticking a box to accept them. ignoring the terms was not enough on its own, and a tick box is not a technical lock.

what is still open. no EU court has said whether a browsewrap in a footer binds a business that scrapes logged out. expect the answer to differ by country, and to depend on what you knew.

what US courts say

the EU rules leave much to national law. US courts, on the other hand, have ruled on these patterns many times, and the names in this post come from their rulings. the main lessons fit in a short list.

• clicks bind, footers often don't. a US appeal court put it this way in 2014: a terms link at the bottom of every page, without more, didn't bind a shopper, because "the onus must be on website owners to put users on notice". in 2002 the same idea sank software terms that showed up only if you scrolled down past the download button.
• a clear sign-up line can be enough. in 2017 another appeal court enforced terms behind a link just below the "register" button. the notice was easy to see, and signing up meant yes.
• knowing the terms changes things. the same 2014 ruling lists cases where courts did enforce footer terms: against a company that sent automated queries every day and admitted it knew the terms, and against a scraper that kept going after a cease-and-desist letter quoted them. a New York court noted in 2015 that footer terms have mostly been enforced against "knowledgeable accessors, such as corporations".
• accounts carry their terms with them. hiQ won the hacking question against LinkedIn, but lost on contract in 2022, because it had used accounts and hired people to make fake profiles.
• logged-out scraping can sit outside the terms. in 2024 a court found that Meta's terms covered what users do on Facebook and Instagram, not scraping of public pages while logged out. Meta dropped the case.

what this means for your project

the web works because reading a public page needs no permission. a site can close that door to you, but it has to say so in a way you can see. the law gives a fairly clear picture for accounts, and an open one for footers. here is how to turn that into a plan, from the easy case to the careful one.

public pages, no account, no click. this is where most projects live: prices, product pages, public notices. at most a browsewrap applies to you. read the terms once anyway: they tell you what the site wants, and that is worth knowing before you build on its data. if they ban scraping, decide with your team whether to go ahead, ask for an api or a license, or skip the site.

keep a note of what you saw. before you start, save a dated copy of the site's terms and robots.txt. if a question comes up a year later, you can show what the page said on the day you began.

pages behind an account. if you or your client opened an account, you clicked through its terms, so treat them as binding. if they ban automated access, don't scrape through that account. ask the site for access instead.

a letter arrives. if a site writes to you and quotes its terms, take it seriously: it is the clearest sign that the site doesn't want your scraper, and US courts have held scrapers to terms after a letter like that. pause, read the letter with your lawyer, and decide before your next run.

the other ways a site says no. besides its terms and a letter, a site can put pages behind a login or a sign-up, block your IP addresses, or close paths in robots.txt. none of these settles the question on its own, and courts weigh them differently. but each one tells you what the site wants, so treat it as an answer, not an obstacle. working around one, with new IP addresses after a block or an account made only to scrape, is where a contract dispute can turn into the kind of case computer crime laws are written for.

every rule has to pass, not just one. the terms are one question. personal data, copyright and the database right still apply, whatever the terms say. and getting past a login or another lock can be a crime, while breaking terms is a contract dispute. a project has to clear all of these; clearing one doesn't clear the others. is web scraping legal in the EU in 2026? covers each one.

expect the law to move. this area changes fast. courts disagree with each other, and now and then the same court changes course in the same case. what holds today may not hold in two years, so check again before a big new project, and decide with your lawyer how much risk your project can carry.

questions people ask

the questions we hear most, with short answers.

this is not legal advice. for a decision on a real project, talk to a lawyer or your DPO.