Hey bro I completely understand why you are checking out this LinkedIn is a goldmine of information, yet on the other hand, they are very protective of it. You can imagine LinkedIn as a club with no door (the public web), but the moment you enter the door (log in), you are expected to abide by their rules about the house.
The name extraction and automation rules here have been reduced to a lowdown that can be easily understood.
1. The Golden Rule of Public vs. Private.
The LinkedIn data are broadly categorized into two buckets. The secret is to understand the distinction between being in trouble and not.
- Public Data (The "Front Window): It is the data you view when not logged in and it typically consists of Full Name, Headline and Location.
- The Rule: Scraping this is technically legal in most locations (with the help of some landmark court cases such as hiQ vs. LinkedIn), but against the rules that LinkedIn designed.
- Private Data (Behind the Login): After logging in, you will have the ability to see a lot more such as mutual connections, full work history, and contact information.
- The Rule: This is a great no-no to be taken out with a tool. As you signed their Terms of Service, when creating your account, it is a breach of contract to do so.
2. The White Hat version of LinkedIn by its definition.
To be 100% compliant, the LinkedIn API is the only official method of extracting names.
- The Boundary: You can drag the names of people who have given your application permission in a deliberate action.
- Fidelity: You cannot simply draw the names of people you are not even acquainted with. The API can be used to such features as Sign in with LinkedIn or management of your profile.
3. Boundaries of Automation (How they will get you)
In case you are using third-party automation systems, the LinkedIn fence has a digital fence that is extremely difficult to cross. The following is what is across the fence:
- High Volume: It is massive alarm bells to pull out over 50-100 profiles per day.
- Bot Behavior: When a tool clicks or scrolls too rapidly (at a faster rate than a human would), the AI of LinkedIn will be signaled.
- Fraudulent Accounts: Fraudulent use of accounts (to scrape names) is violating the policy and normally results in the IP being blocked.
4. Step by Step Compliance Checklist.
To be safe when you are constructing or operating a tool or using it, do the following:
- Use Public URLs: Only visit publicly accessible accounts in order to avoid breaking and entering (legal-wise).
- Use a Human Pace: Random delays are to be used in your automation tool. Do not scrape 100 profiles in 5 minutes; do it all throughout the day.
- Respect the robots.txt: This is a document on their site which informs the bots where they can and cannot access. This is automatically checked with good tools.
- GDPR/Privacy Check: Although it is possible to extract a name, in case that individual is located in the EU or California, then you may need a legal purpose to store their information. Always provide a means through which people can request you to delete their info.
- Don't Touch Emails: It is one thing that names are taken, but scraping personal emails is where LinkedIn (and the law) gets very aggressive.
The Bottom Line
The official position of LinkedIn is: Don't do it. They do not want the presence of bots in their site. When you take names with a tool you are always playing a little game with your account.
When you are simply performing a small amount of data processing on a project, you can afford to keep it slow. However, to get to the level of what one can call industrial-scale extraction, you are most certainly going to cross the policy boundary.
Wish that will make you straighten things out! Get in touch with me in case you require assistance with a particular tool or another problem.
