Is LinkedIn scraping legal? Not by LinkedIn's rules (2026)
2026-11-16 · updated 2026-11-16 · 13 min read · Adam Axelsson, founder of Revexa
Scraping LinkedIn breaks LinkedIn's rules, and courts have enforced them: section 8.2 of the User Agreement bans software "to scrape or copy the Services, including profiles and other data from the Services". In the US, the Ninth Circuit held in hiQ v. LinkedIn (2022) that "without authorization" in the federal anti-hacking law "does not apply to public websites", but hiQ still lost on contract: the court found that the User Agreement "unambiguously prohibits" its scraping, and the case ended in December 2022 with a $500,000 judgment and a permanent ban. In the EU, scraped profiles are personal data under GDPR: France's CNIL fined Kaspr €240,000 in 2024, in particular for collecting LinkedIn contact details that users had restricted to their connections. Using LinkedIn normally, by viewing profiles and messaging people yourself, is not scraping.
Checked on 16 November 2026 against LinkedIn's User Agreement and help pages, the court records in hiQ Labs v. LinkedIn, the Supreme Court docket, LinkedIn's own announcements, and the decisions of the CNIL, the Spanish AEPD and the Irish DPC. This is not legal advice.
Revexa builds Leo, a tool that reaches people on LinkedIn, so read with that in mind. For the rules on automating your own account, see is LinkedIn automation legal and LinkedIn outreach without getting restricted. For scraping other websites, see is web scraping legal for lead generation.
LinkedIn scraping at a glance
| Situation | Rule | Source |
|---|---|---|
| Viewing profiles and messaging people yourself | Normal use; LinkedIn's limits still apply | User Agreement 3.4 |
| Exporting your own 1st-degree connections | Allowed through LinkedIn's own export; emails only where each connection allows it | LinkedIn Help |
| A bot or browser extension that copies profiles | Banned by the User Agreement | Section 8.2; Prohibited software page |
| Crawling LinkedIn without permission | "strictly prohibited"; permission is for search indexing | Crawling Terms |
| Fake accounts to reach logged-in data | Banned; fake accounts featured in hiQ, Proxycurl and ProAPIs | Section 8.2; court records; LinkedIn |
| Federal hacking law (CFAA), public profiles | "Without authorization" does not apply to public websites (Ninth Circuit) | hiQ v. LinkedIn, 2022 |
| Breach of LinkedIn's contract | Enforced: hiQ agreed to a $500,000 consent judgment | N.D. Cal., 8 December 2022 |
| Scraped personal data in the EU | GDPR applies: a legal basis, informing people of the source, the right to object | GDPR art. 6, 14, 21 |
| Maximum fine under GDPR | €20 million or 4% of worldwide turnover, whichever is higher | GDPR art. 83(5) |
Scraping versus using LinkedIn normally
Scraping means collecting LinkedIn's pages or data with software: a crawler, a script, a scraping API or a browser extension that reads profiles and stores them. The User Agreement bans that whether the tool runs in the cloud or in your browser.
Normal use is what LinkedIn is built for: searching, viewing profiles, sending invitations and messages yourself, using Sales Navigator inside LinkedIn, and downloading your own data. LinkedIn's help center says "You can export a list of the connections you have on LinkedIn at any time", with two limits: "You will only see email addresses for connections who have allowed their connections to see or download their email address", and "You currently can't export a list of your contacts that aren't 1st-degree connections" (Export connections).
Two points blur the line:
- Copying by hand. Section 8.2 also bans copying or using "any information (including content) obtained from the Services" without the content owner's consent. We found no LinkedIn guidance on a salesperson noting one prospect's name and title in a CRM. Our reading: that is ordinary use, while building a list or database from profiles is what the clause targets.
- The message you send afterwards. Even a profile you looked up by hand is personal data, and the email you send from it is governed by marketing law. In Spain, the recruiter Hays looked up a person's LinkedIn profile and sent him a marketing email without consent; the AEPD proposed a €5,000 fine under the Spanish e-commerce law (LSSI article 21), and Hays paid €4,000 on 4 January 2023 to close the case (AEPD, EXP202209558). Country rules: cold email laws by country.
What LinkedIn's User Agreement says
The User Agreement, in force since 3 November 2025, is a contract you accept "By creating a LinkedIn account or accessing or using our Services". In section 8.2 you agree that you will not:
"Develop, support or use software, devices, scripts, robots or any other means or processes (such as crawlers, browser plugins and add-ons or any other technology) to scrape or copy the Services, including profiles and other data from the Services"
"Override any security feature or bypass or circumvent any access controls or use limits of the Services (such as search results, profiles, or videos)"
"Copy, use, display or distribute any information (including content) obtained from the Services, whether directly or through third parties (such as search tools or data aggregators or brokers), without the consent of the content owner (such as LinkedIn for content it owns)"
"Rent, lease, loan, trade, sell/re-sell or otherwise monetize the Services or related data or access to the same, without LinkedIn's consent"
"Use bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages"
The same section bans creating "a false identity on LinkedIn". Under section 3.4, "LinkedIn reserves the right to restrict, suspend, or terminate your account if you breach this Contract or the law or are misusing the Services". Section 5 lists section 8.2 among the terms that survive when the contract ends.
LinkedIn's help center applies this to tools: it does not permit third-party software, including crawlers, bots and browser extensions, "that scrape, modify the appearance of, or automate activity on LinkedIn's website" (Prohibited software and extensions). Its separate Crawling Terms, last revised 25 May 2017, say: "Automated Crawling & Indexing without the express permission of LinkedIn is strictly prohibited." Permitted crawlers may use the data only for "search indexing for display in a publicly available search engine on the Internet unless granted separate approval by LinkedIn".
Who is bound? hiQ did not dispute that it was: on LinkedIn's contract claim, "hiQ only contests the breach and damages elements" (summary judgment order, October 2022). Whether a scraper who never logs in and never accepts the terms is bound by them is contested, and we found no ruling on that point for LinkedIn. For Meta's terms, the same judge held in January 2024 that they did not apply to Bright Data's logged-off scraping of publicly viewable data (see is web scraping legal for lead generation). The hiQ injunction covered automated copying "whether logged in to a LinkedIn account or not", but hiQ agreed to that wording.
hiQ v. LinkedIn: the full history
hiQ Labs sold analytics built from public LinkedIn profiles. After LinkedIn's cease and desist letter, hiQ sued first, and the case ran for five and a half years.
| Date | What happened | Source |
|---|---|---|
| 23 May 2017 | LinkedIn sends hiQ a cease and desist letter citing its User Agreement and federal and state law | Stipulation, Dkt. 405 |
| June 2017 | hiQ sues LinkedIn in the Northern District of California (No. 17-cv-03301-EMC) | Dkt. 405 |
| 14 August 2017 | Judge Edward Chen grants hiQ a preliminary injunction: LinkedIn may not block hiQ's access to public profiles | Order, Dkt. 63 |
| 9 September 2019 | The Ninth Circuit affirms the injunction | Supreme Court docket 19-1116; 2022 opinion |
| 14 June 2021 | The Supreme Court vacates and remands "for further consideration in light of Van Buren v. United States" | Docket 19-1116 |
| 18 April 2022 | The Ninth Circuit affirms again: "the concept of 'without authorization' does not apply to public websites" | Opinion |
| 27 October 2022 | Summary judgment order (filed publicly 4 November 2022): the User Agreement "unambiguously prohibits hiQ's scraping and unauthorized use of the scraped data" | Order, Dkt. 404 |
| 6 December 2022 | The parties file a stipulated consent judgment as a condition of a confidential settlement | Dkt. 405 |
| 8 December 2022 | Judgment of $500,000 against hiQ and a permanent injunction | Consent judgment, Dkt. 406 |
What the Ninth Circuit decided
The 2022 opinion reviewed a preliminary injunction, so it decided only that hiQ had raised serious questions, not who would win. On the Computer Fraud and Abuse Act (CFAA) it followed the Supreme Court's "gates-up-or-down" test in Van Buren: public pages have no gate to lift, so "Van Buren therefore reinforces our conclusion that the concept of 'without authorization' does not apply to public websites." It added that scraping victims "are not without resort, even if the CFAA does not apply", naming trespass to chattels, copyright infringement, misappropriation, unjust enrichment, conversion, "breach of contract, or breach of privacy".
What the district court decided in 2022
Back in the district court, LinkedIn won on contract. The October 2022 order found that "hiQ breached LinkedIn's User Agreement both through its own scraping of LinkedIn's site and using scraped data, and through turkers' creation of false identities on LinkedIn's platform." The turkers were independent contractors hiQ directed to make accounts and copy profile URLs, and hiQ's own instructions told them: "It is a good idea to make a fake account with a fake email, to deal with the possibility of being banned on LinkedIn."
The court granted summary judgment only on the fake accounts. On the scraping itself, the court left hiQ's waiver and estoppel defenses for trial. The same order threw out hiQ's own unfair competition and interference claims under California's litigation privilege.
How it ended
There was no trial. In the stipulation, hiQ accepted that LinkedIn could establish liability under the CFAA "based on hiQ's data collection practices and based on hiQ's direct access to password-protected pages on LinkedIn's platforms using fake accounts", plus liability under California Penal Code § 502 and for trespass to chattels and misappropriation. The consent judgment of 8 December 2022 ordered:
- "Judgment in the amount of $500,000 USD is hereby entered against hiQ and in favor of LinkedIn."
- A permanent ban on "using automated means to access and/or copy data from the LinkedIn platform, whether logged in to a LinkedIn account or not, without express written permission of LinkedIn", on fake accounts, and on using LinkedIn "to develop a commercial service" without permission.
- Deletion of all hiQ code built to access LinkedIn or to use data from it, and of all LinkedIn member profile data it held.
What hiQ means today: scraping public LinkedIn pages is probably not federal hacking in the Ninth Circuit, but it is a breach of contract that LinkedIn can sue over, and fake accounts or logged-in scraping add CFAA and state computer-crime exposure. The CFAA concessions were agreed by hiQ, not decided by the court, so they bind hiQ and set no precedent.
LinkedIn's lawsuits against scrapers, 2024 to 2026
We found no new LinkedIn scraping lawsuit filed in 2024. These are the cases we could confirm from 2025 and 2026:
| Date | What happened | Source |
|---|---|---|
| 24 January 2025 | LinkedIn files a federal lawsuit against Proxycurl (Nubela) and its founder for "the unauthorized creation of hundreds of thousands of fake accounts and scraping of millions of LinkedIn member profiles" | Sarah Wight, LinkedIn VP of Legal; LinkedIn Pressroom |
| 4 July 2025 | Proxycurl shuts down. Its CEO: "Regardless of the merits of LinkedIn's lawsuit, there is no winning in fighting this", and the company closed "to comply with the legal settlement with LinkedIn" | Nubela |
| 28 July 2025 | LinkedIn says Proxycurl must delete all LinkedIn data obtained through unauthorized means and stop accessing LinkedIn unlawfully: "The Court has entered these requirements as a permanent injunction, which Proxycurl is obligated to send to its customers." The court docket (N.D. Cal. No. 3:25-cv-00828) shows a judgment entered on 25 July 2025 | Sarah Wight; docket |
| 2 October 2025 | LinkedIn sues ProAPIs and its principal for scraping "millions of LinkedIn member profiles, including by using over a million continuously created (and swiftly restricted) fake accounts". The complaint alleges that customers paid "up to $15,000 per month to scrape LinkedIn" | Sarah Wight; complaint, N.D. Cal. No. 5:25-cv-08393 |
| 6 April 2026 | In a case Teamfluence brought, LinkedIn says a German court rejected Teamfluence's request to restore accounts LinkedIn had restricted for a scraping browser extension, and "found Teamfluence itself is violating data protection laws". Teamfluence says the Munich Regional Court (37 O 104/26) "found that Teamfluence's use of data from users who authorized data export is permissible", declined the injunction, and that the decision is on appeal. We found no published decision | Sarah Wight; Teamfluence, 1 July 2026 |
| 16 September 2026 | Judge P. Casey Pitts signs a final judgment on consent: ProAPIs, Netswift (SMC-Private) Limited and their principal are permanently barred from accessing LinkedIn through fake accounts, scraping, crawling or bots and from transferring LinkedIn data; they deny liability and certified that they deleted all LinkedIn data. LinkedIn: "ProAPIs must stop scraping LinkedIn, selling or transferring LinkedIn data, accessing LinkedIn through fake accounts, and it must delete the LinkedIn data it holds." ProAPIs said on its website that it "does not offer tools to scrape LinkedIn" | Judgment on consent, Dkt. 36; Sarah Wight; The Record, 21 September 2026 |
The pattern: both US cases since 2025 centred on fake accounts, which in our reading keeps them away from the hiQ ruling on public pages, and both ended with the scraper bound to stop and delete the data. LinkedIn says every lawsuit it has filed against scrapers has ended in a judgment prohibiting scraping (Sarah Wight, 2 October 2025). Both outcomes were agreed: ProAPIs consented to judgment while denying liability, and Proxycurl's injunction followed a settlement, so no court ruled on the merits.
GDPR: scraped LinkedIn data is still personal data
A public profile is not free to use in the EU. Twelve data protection authorities, among them the UK's ICO, Canada's OPC, Norway's Datatilsynet and Switzerland's FDPIC, said in a joint statement sent to LinkedIn's owner Microsoft and other platforms: "Personal information that is publicly accessible is still subject to data protection and privacy laws in most jurisdictions." Among the risks they list: "Unwanted direct marketing or spam" (joint statement, 24 August 2023).
If you scrape or buy LinkedIn data about people in the EU, GDPR (as adopted) requires:
- A legal basis under article 6, usually legitimate interest with a documented balancing test. More in GDPR and B2B email under legitimate interest.
- Telling people where you got their data. Article 14(2)(f) requires the source, "and if applicable, whether it came from publicly accessible sources", and article 14(3)(b) sets the deadline "at the latest at the time of the first communication to that data subject".
- Stopping at an objection. "Where the data subject objects to processing for direct marketing purposes, the personal data shall no longer be processed for such purposes" (article 21(3)).
The ceiling for breaching these articles is "20 000 000 EUR, or in the case of an undertaking, up to 4 % of the total worldwide annual turnover of the preceding financial year, whichever is higher" (article 83(5)).
The cases we found
- Kaspr, France (5 December 2024): €240,000. Kaspr's Chrome extension showed contact details of people whose LinkedIn profiles its customers visited, from a database of "about 160 million contacts". The CNIL fined it "in particular because it collected contact details of users on LinkedIn, even if they previously masked them": collecting details people had limited to their 1st and 2nd-degree connections "exceeded what could reasonably be expected from people who register on a professional social network". It also found excessive retention, late information sent only in English, and incomplete answers about the data's source (CNIL). On 4 March 2026 the CNIL closed its order after Kaspr "chose to delete its database and cease all data collection on LinkedIn" (CNIL).
- Hays, Spain (2023): €4,000 paid. The profile lookup and marketing email described above, sanctioned under Spain's ePrivacy rules rather than GDPR (AEPD).
Note what the CNIL did not say: it called the details of people who left them visible to everyone "data collected by the company in a lawful manner". For those data it faulted the retention period, and its findings on late information and incomplete source answers applied to everyone. Our reading: public visibility can support a legal basis; it does not remove the duties to inform, limit retention and honour objections.
Not a scraping case: the Irish DPC's €310 million fine on LinkedIn on 24 October 2024 concerned LinkedIn's own processing of members' data for behavioural analysis and targeted advertising (DPC).
Buying scraped LinkedIn data
Buying data does not move the risk to the vendor. Section 8.2 bans using information obtained from LinkedIn "whether directly or through third parties (such as search tools or data aggregators or brokers)" without consent. Under GDPR you become a controller of what you buy and must name the source within one month of getting the data, and at the latest in your first message (article 14(3)). And when LinkedIn wins, the data can disappear: Proxycurl had to send its injunction to its customers, and ProAPIs agreed to stop selling or transferring LinkedIn data.
Our reading: before buying contact data, ask the vendor in writing where the data comes from and whether it comes from LinkedIn profiles, and keep the answer.
Checklist
- Do not run scrapers, scraping APIs or extensions on LinkedIn. Section 8.2 bans them, and LinkedIn restricts accounts that use them.
- Never use fake or shared accounts. Fake accounts were central in hiQ, Proxycurl and ProAPIs.
- Use LinkedIn's own tools: search, Sales Navigator and the export of your own connections.
- Ask data vendors where their data comes from before you buy, and get it in writing.
- For EU contacts, document a legal basis, tell people the source within one month and at the latest in your first message, and stop at the first objection.
- Check the marketing law of the recipient's country before you email someone you found on LinkedIn: cold email laws by country, for example the US, the UK, France and Spain.
Where Leo fits
Leo, the buying-signal agent Revexa builds, reaches people on LinkedIn through the customer's own profile and by email through a connected mailbox. In review mode, the default, the customer reads and approves every draft before it goes out, and whoever sends a message is responsible for following the law where the recipient is. Ways to work LinkedIn by hand or with tools: best LinkedIn automation tools 2026.
Common questions
Is scraping LinkedIn illegal?
It breaks LinkedIn's User Agreement, which courts have enforced: hiQ agreed to a $500,000 consent judgment in 2022. In the US, scraping public profiles is probably not a federal hacking crime after hiQ v. LinkedIn, but fake accounts and logged-in scraping add risk. In the EU, GDPR applies to the scraped data.
Did hiQ win against LinkedIn?
It won the early rounds on the federal hacking law: in 2022 the Ninth Circuit, reviewing a preliminary injunction, held that "without authorization" does not apply to public websites. It lost the case: in 2022 the district court found it breached the User Agreement, and hiQ agreed to a $500,000 judgment and a permanent ban on scraping LinkedIn.
Is scraping public LinkedIn profiles legal under GDPR?
Not automatically. Public data is still personal data. You need a legal basis, must tell people where you got their data within one month and at the latest in your first message, and must stop at an objection. The CNIL fined Kaspr €240,000, in particular for collecting contact details people had restricted to their connections.
Why did Proxycurl shut down?
LinkedIn sued it in January 2025 over hundreds of thousands of fake accounts used to scrape millions of profiles. Proxycurl closed on 4 July 2025 "to comply with the legal settlement with LinkedIn", and LinkedIn says a permanent injunction requires it to delete all LinkedIn data obtained through unauthorized means.
Is exporting my LinkedIn connections scraping?
No. LinkedIn's own export lets you download your 1st-degree connections, with email addresses only for those who allow it. Using a tool to collect profiles beyond that is scraping.
Can I buy LinkedIn data from a vendor?
LinkedIn's User Agreement bans using its data obtained through "data aggregators or brokers" without consent, and under GDPR you must name the source to the people concerned. Ask where the data comes from before you buy.