TL;DR
The notice arrives at the worst possible time. An SDR is mid-sequence with 400 prospects in play, and LinkedIn says the account is restricted. What happens in the next hour decides whether the account is back in a week or gone for good. Most of the advice online is written by automation vendors with a tool to sell, and it blurs five different restrictions into one generic "LinkedIn jail". This guide separates them, quotes what LinkedIn actually publishes for each, and lays out a recovery sequence that a team running LinkedIn plus cold email can execute without making the situation worse.
Sourcing note: quotes attributed to LinkedIn come from its Help Center, User Agreement (effective 3 November 2025) and Community Report, fetched in September 2026. Durations, thresholds and recovery ramps from automation vendors and community posts are marked (directional): they are observed, not official, and LinkedIn Support "cannot disclose the type or reason for the restriction", so nobody outside LinkedIn can verify them.
A LinkedIn restriction is not one thing. The five that hit outbound teams are: a temporary invitation restriction (lifts on its own, usually within a week), an automated-activity suspension (lifts at the date printed on the notice once the tool is disabled), an identity-verification lock (lifts only after a Persona ID check), a content or profile restriction (appealed through on-screen prompts), and a permanent restriction (LinkedIn's only published path is "contact us"). The commercial use limit on search is a sixth thing people confuse with a restriction, and it simply resets on the first of the month.
The recovery sequence is the same for all of them in the first hour: disconnect every tool and browser extension, read the exact wording of the notice, complete any on-screen security step, and stop all outbound on that account. After that, the paths diverge, and the rest of this guide follows each one. If you are here for prevention rather than repair, our LinkedIn limits and account safety guide covers the volumes, acceptance rates and warm-up ramps that keep you out of this article in the first place.
LinkedIn's Account restrictions help article groups restrictions under four headings: content violations, profile violations, identity violations and automated-tools violations. Its invitation-restriction articles add a fifth, and its commercial-use article describes a sixth that is technically a limit rather than a restriction. Here is how they differ, with LinkedIn's own wording where it exists.
| Restriction | Official trigger | What LinkedIn publishes | Official way out |
|---|---|---|---|
| Invitation restriction (temporary) | "Many invitations within a short amount of time"; invitations "ignored, left pending, or marked as spam"; too many "I don't know this person" responses | "Typically lasts one week"; "LinkedIn cannot remove or shorten the wait period" | Wait. First restriction: "a few hours". Multiple in a day: "a few days". Pending pile-up: "up to one month" |
| Automated-activity suspension | Third-party software or extensions that "scrape, modify the appearance of, or automate activity" | "Automatically be re-enabled at the time specified in the suspension notice"; "repeated suspensions may result in permanent restriction" | "Review and disable the software or extension", then wait for the stated date. Optional contact form |
| Identity-verification lock | Profile "intentionally fraudulent"; "signs that your account has been compromised"; profile or identity restriction you believe is an error | "Access to your account may be restricted either temporarily or indefinitely" | "Follow the onscreen steps to verify your identity" via Persona with a government ID |
| Content or profile restriction | Professional Community Policies breach; non-compliant name, photo, experience or education | Some violations "may result in permanent account restriction after a single violation" | "Follow the onscreen prompts to ask us to revisit our decision"; LinkedIn reviews and reports the outcome |
| Permanent restriction | Egregious violations or repeat offences | "You will not be able to use your account... Other people will no longer be able to find your LinkedIn profile or message you" | "Contact us". No reversal path is published |
| Commercial use limit (search only) | Activity that "indicates that you're likely using LinkedIn for commercial use, like hiring or prospecting" | "We are not able to display the exact number of searches or views you have left" | Resets "at midnight PST on the 1st of each calendar month", or upgrade to Premium Business, Recruiter Lite or Sales Navigator |
The single most useful thing you can do in the first minute is match your notice to a row. The wording differs: "temporarily restricted from sending invitations" is row one, "we've detected unusual activity" or "we need to verify your identity" is row three, and a notice with a specific re-enable date is row two. Automation vendors have catalogued the message variants (directional), but the notice itself is the only authority on which restriction you have.
This is the restriction most outbound accounts hit first, and it is the one with the most misinformation around it. LinkedIn's Invitation limit reached article is unusually direct. During the restriction, "you will not be able to send invitations until the restriction is lifted", "withdrawing pending invitations will not remove the restriction", "you cannot buy or acquire more invitations while restricted", and "LinkedIn cannot remove or shorten the wait period." The Types of restrictions for sending invitations article adds that "LinkedIn won't be able to remove invitation restrictions upon request" and gives the only official timing guidance that exists:
Three official triggers are named: volume in a short window, invitations that were ignored, left pending or marked as spam, and volume that makes LinkedIn "suspect the use of an automation tool". A separate article confirms that "if someone receives an excessive number of I don't know this person responses, they could be restricted from sending invitations", and that senders cannot see who marked them or how many did. Note what is absent: LinkedIn publishes no number for the weekly invitation limit anywhere. The "about 100 a week" figure that every vendor repeats (directional) is consistent across sources, but it is observed, not official.
Why the "withdraw your invites to get unrestricted" advice persists. Withdrawing stale invitations does not lift a restriction and does not refund quota. It does address one of the three official triggers (invitations "left pending"), and vendors argue it improves the acceptance-rate signal on the account (directional). So it is a sensible clean-up during the wait, not a key that opens the door. Two official side effects to know: the recipient "will not be notified", and "you won't be able to send a new invitation to the same member for up to three weeks." Invitations expire on their own after six months.
LinkedIn's position on automation is not ambiguous, and the recovery instruction follows from it. The Automated activity on LinkedIn article says: "If your account is restricted for automated activity, we ask that you take the following actions: review and disable the software or extension that automates activities. Your account will then automatically be re-enabled at the time specified on the suspension notification." The invitation article adds the escalation clause: "Repeated suspensions may result in permanent restriction of your LinkedIn account."
Read together, that means the suspension is on a timer you cannot shorten, and the only variable you control is whether it happens again. A tool that keeps logging in during the suspension is the one thing that can convert a temporary hold into a permanent one. The User Agreement (Section 8.2, the "Don'ts") is what the notice is enforcing. The relevant lines prohibit using "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", using "bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages", and any attempt to "override any security feature or bypass or circumvent any access controls or use limits". The Prohibited software and extensions article warns that such tools "may become non-operational without notice", and the automation article notes their use "may be a violation of privacy legislation in specific jurisdictions."
Two practical implications for teams that use automation anyway, which is most of the teams reading this. First, every tool on the market violates these terms, whichever architecture it uses. Vendors selling cloud tools with dedicated IPs say browser extensions are the riskiest category (directional, and self-serving), and two competitor sources report that LinkedIn removed a well-known cloud tool's company page in March 2026 (directional, secondary sources only). Treat architecture as risk reduction, never immunity. Second, LinkedIn's own Commercial use limit reset article observes that plug-ins can "run searches and view profiles in the background, which can cause you to surpass the limit without actually seeing any of the incremental warnings." A tool can be spending your account's budget while you are not looking. Our LinkedIn automation tools comparison covers how the main tools handle pacing and limits.
When the notice says LinkedIn needs to verify your identity, you are in the one restriction that does not lift on a timer. LinkedIn's Verify your identity to recover account access article (updated in late August 2026) describes the flow. Persona, "LinkedIn's identity verification provider", asks for "a clear photo of your valid government issued ID card, driver's license, or passport", and "in some instances, Persona may ask you to take a photo of yourself to verify that your face matches your ID portrait." Library and school ID cards are not accepted, and accepted documents "may vary by country". Persona shares only the verification result, full name, year of birth, city and country, ID type and issuer, and a redacted copy of the ID with the LinkedIn account; LinkedIn "doesn't receive your biometric data or numbers, expiry dates, or issue dates", and the data is "generally permanently deleted within 14 days of submission." The article also references "other ways to recover your account" for members who do not want to submit an ID.
The name on the ID must match the name on the profile. That is the detail that sinks outbound accounts most often: a profile using a nickname, a maiden name, or a slightly anglicised spelling will fail an automated match. The same rule applies to LinkedIn's voluntary verification badge, which runs through CLEAR in the US, Canada and Mexico, Persona in many other countries (NFC-enabled passports), and DigiLocker in India. LinkedIn passed 100 million members with a verification in December 2025. Notably, LinkedIn makes no claim anywhere that a verified profile is less likely to be restricted. The published benefits are more profile views and engagement, and CLEAR-verified members are enrolled in two-factor authentication.
| Question | What LinkedIn publishes | What vendors and users report (directional) |
|---|---|---|
| How long is an invitation restriction? | "Typically lasts one week"; hours for a first hit, up to a month for a pending pile-up | Vendors most often say 24 to 72 hours for a first restriction, a few days to a week for repeats |
| How long is an automation suspension? | Until "the time specified in the suspension notice" | One vendor quotes the notice as "restricted for 1/24 hours" for a first offence; a second warning quotes "we have now restricted your account until [date]" |
| How long does the ID check take? | No timeline published anywhere | Vendor guides: 1 to 3 days, 24 to 72 hours, or 3 to 14 days depending on who you read. First-person accounts: 7 to 10 business days after escalation, about a month when Persona misread an ID, and roughly six months in one case involving a previously deleted account |
| Can support tell me why? | "LinkedIn Support cannot disclose the type or reason for the restriction" | Consistent with every user report we found |
| Can a permanent restriction be lifted? | Not stated either way; "contact us" is the only instruction | Vendor "appeal success rate" tables exist, but the source itself says they are not official figures and gives no sample, so we do not repeat them |
The spread in the ID-check column is the honest picture. The fast cases are people whose ID matched cleanly and who submitted once. The slow cases involve an automated document reader rejecting a valid ID, and the one person who fixed it did so by resubmitting with a clean photo after covering a field the reader was misparsing (directional, single report). Submit one clear, well-lit, matching ID and then wait. Do not resubmit daily. The User Agreement separately prohibits members who "misuse our reporting or appeals process, including by submitting duplicative, fraudulent or unfounded reports, complaints or appeals", so a flood of appeals is itself a violation.
The Permanent account restrictions article is short: "You will not be able to use your account or any of LinkedIn's features associated with your account. Other people will no longer be able to find your LinkedIn profile or message you." The only action it offers is "contact us". It does not say the decision can be reversed, and it does not say it cannot.
What it also does not say, but the User Agreement does, is that opening a replacement account is a fresh violation. Section 2.2 requires that "you will only have one LinkedIn account, which must be in your real name", and that you are "not already restricted by LinkedIn from using the Services" when you register. The Merge or close duplicate accounts article repeats that "creating multiple profiles is not allowed." Every practitioner source we checked, including the vendors whose business depends on people keeping accounts alive, says the same thing about what happens next: the new account gets restricted and the old one is flagged for good (directional). The mechanism reported is linkage by device fingerprint, IP and the identity on the government ID (directional); the one long first-person account we found described a new profile tied to the same ID as a deleted one being shut down repeatedly for about six months.
Scale of enforcement, from LinkedIn's own Community Report (July to December 2025): automated defences "blocked 97.8% of the fake accounts we stopped", "99.7% of the fake accounts were stopped proactively, before a member report", and LinkedIn is deploying "improved models that are restricting more fake accounts at registration before they have a chance to share content." The absolute counts sit in charts we could not render, so we are not quoting a headline number, but the direction is clear: a new account created after a ban is being screened at registration by systems tuned to catch exactly that.
If an SDR's account is permanently restricted, the practical answer is that the person's LinkedIn presence is gone and the team's outbound has to be re-planned around cold email and the remaining accounts. That is a hard message, and it is why the rest of this guide focuses on not getting there.
The sequence below merges LinkedIn's instructions with the vendor guidance that is consistent across sources. Where the vendors disagree, we say so. Where a step is only supported by a single user report, it is marked.
| Step | When | Why | Basis |
|---|---|---|---|
| 1. Disconnect every automation tool and extension; sign out of all sessions; clear cookies | Hour 0 | LinkedIn's stated first step. A tool that keeps logging in extends the flag and risks the "repeated suspensions" clause | Official |
| 2. Screenshot the notice and match it to the table above | Hour 0 | Support will not tell you the reason later. The notice is your only record of type and, for automation, the re-enable date | Official |
| 3. Complete the on-screen security step (CAPTCHA, phone, password reset if a new-location warning was involved) | Hour 0 to 1 | Standard unlock step; a compromised-account flag needs a password change anyway | Official |
| 4. If asked, submit one clear government ID that matches the profile name exactly | Day 0 to 1 | Mismatched or blurry documents restart the clock; repeated appeals are themselves a violation | Official + user reports |
| 5. Full outbound pause on the account | 48 to 72 hours minimum; most vendors say 3 to 7 days (directional) | LinkedIn's own guidance escalates from "a few hours" to "a few days" for repeat hits | Official + vendors |
| 6. Withdraw stale pending invitations in small batches | Days 1 to 7, invites older than 2 to 3 weeks | Removes an official trigger. Does not lift the restriction or refund quota. Three-week re-invite lockout applies | Official + vendors |
| 7. Pause the account's sequences in your tool and CRM; move live conversations to email | Day 1 | Some tools retry failed sends automatically rather than pausing (directional); a queue of retries against a locked account is a bad look | Vendors |
| 8. Appeal once, if not lifted by the stated date: factual, short, committing to stay within the limits | Day 3 to 7 | "Follow the onscreen prompts to ask us to revisit our decision"; the appeal form is sign-in gated | Official |
| 9. Re-warm manually before reconnecting any tool | Weeks 1 to 4 after the lift | Treat the account as new; see the ramp below | Vendors |
| 10. Never create a second account | Always | One-account rule in the User Agreement; consistent reports of both accounts being lost | Official + vendors + users |
One escalation route deserves a mention with the right caveat. A single 2023 user report describes weeks of silence from support followed by restoration 7 to 10 business days after filing a complaint with a US consumer body, and notes it was a commonly shared tip on forums (directional, single report). We include it because it is the only escalation path anyone reports working, not because it is reliable.
The vendor guides disagree on almost everything except this: after a restriction, the account should behave like a new one for several weeks, and automation should come back last and at a fraction of the previous cap. The ramp below is the overlap between four vendor guides (directional throughout). One vendor publishes a noticeably faster ramp of 15 to 20 invitations a day from the first week; we have left it out because it is an outlier and the downside of being wrong is the "repeated suspensions" clause.
| Week after the lift | Invitations | Other activity | Automation |
|---|---|---|---|
| During the restriction | 0 | Light, human engagement only: a handful of comments and likes a day | Disconnected |
| Week 1 | 5 to 10 a day, manual, personalised, to people with mutual connections | 10 to 20 total actions a day; one profile edit a day at most | Disconnected |
| Week 2 | 5 to 10 a day | Messages to new connections 10 to 20 a day | Reconnect only if fully lifted, at about a quarter of the previous cap |
| Weeks 3 to 4 | 10 to 20 a day, under roughly 100 a week rolling | Start posting two or three times a week | Low volume, consistent device and session, randomised timing |
| Weeks 5 to 8 | Hold at 20 to 25 a day | Keep pending invitations under 300 to 500; withdraw anything older than three weeks | Hold caps |
| Weeks 9 to 12 | Scale only if acceptance rate holds above 25 to 40 percent | Acceptance rate is the governor, not the weekly cap | Return to target caps |
The acceptance-rate line is where the vendors disagree most, with danger thresholds published at 25, 30 and 40 percent (directional). The one survey with a disclosed sample, 104 B2B sellers polled in late 2025, found that sending fewer than 25 requests a week roughly doubled the odds of a 40 percent acceptance rate, and that warming sequences lifted median acceptance from about 15 to about 40 percent (directional, small self-selected sample). Whatever threshold you pick, the lesson is that the fix for a low acceptance rate is tighter targeting and a better reason to connect, which is a copy and list problem. Our LinkedIn message templates post covers the connection-request side of that.
The restriction is an account problem, not a pipeline problem, unless you let it become one. A team that runs LinkedIn as one of two channels barely notices a week-long lock. A team that runs LinkedIn only loses a week of pipeline and then, in panic, does one of the things that turns a temporary hold into a permanent one. Three rules:
For teams that have burned a cold email domain, the logic will feel familiar: it is the same repair-or-retire decision we lay out in the domain blacklist recovery playbook. The difference on LinkedIn is that there is no "retire and buy a new one". The account is the person, and the User Agreement makes sure of it.
Everything above is damage control. The architecture that avoids it is in our LinkedIn limits and account safety guide: a warm-up ramp for every new or dormant account, acceptance rate as the governor rather than the weekly cap, conservative daily volumes with randomised timing, one device and one session per account, and pending invitations kept low and cleaned weekly. Two additions from this research. First, keep the profile name identical to the name on a government ID, because the Persona check is the one gate you cannot talk your way through. Second, decide in advance what the team does on the day an account is restricted, so nobody improvises a second account or a shared login at 5pm on a Friday.
It is also worth being honest about the terms. LinkedIn's help centre invites members who use third-party tools "for productivity reasons" to submit a form so LinkedIn can "consider partnering with these applications", which is the closest LinkedIn comes to acknowledging that the tools exist. It is not permission. Any team automating LinkedIn is operating against the User Agreement, and the sensible way to hold that position is with volumes low enough that the account would look normal without the tool. That is how we run it for clients, and it is why LinkedIn is never the only channel in a GenFlows build.
It depends on the type. LinkedIn says an invitation restriction "typically lasts one week", with a first restriction clearing in "a few hours" and a pending-invitation pile-up taking "up to one month". An automation suspension lifts "at the time specified in the suspension notice". An identity-verification lock has no published timeline and lifts only after a Persona ID check, which vendor guides put at anywhere from one to fourteen days (directional). A permanent restriction has no published end.
No. LinkedIn's help centre states that "withdrawing pending invitations will not remove the restriction" and that it "cannot remove or shorten the wait period." Withdrawing stale invitations is still worth doing during the wait, because too many pending invitations is one of the three official triggers, but it does not open the door and you cannot re-invite the same person for up to three weeks.
Log in and follow the on-screen prompts. For content and profile decisions LinkedIn says you can "ask us to revisit our decision" and it will report the outcome. For identity or compromised-account restrictions you complete an identity check through Persona with a government ID that matches your profile name. Submit one clear document and wait; the User Agreement prohibits "duplicative, fraudulent or unfounded" appeals, and LinkedIn publishes no response-time commitment.
Not without breaking the User Agreement. Section 2.2 says "you will only have one LinkedIn account, which must be in your real name" and that you must not be "already restricted by LinkedIn" when you register. Practitioner reports consistently describe the new account being restricted and the old one flagged permanently (directional), and the identity check ties both to the same government ID.
Pause that account's LinkedIn sequences, move the live prospects to cold email and phone, and do not run them through a colleague's login, because account sharing is itself prohibited. When the restriction lifts, re-warm the account manually at 5 to 10 invitations a day for two weeks before reconnecting any tool at a fraction of its previous cap (directional).
No tool prevents them. LinkedIn's terms prohibit all third-party software that automates activity, regardless of architecture. Vendors selling cloud tools say browser extensions are the riskiest category and dedicated IPs are safer (directional and self-serving), while competitor sources report a cloud-based tool's company page being removed by LinkedIn in 2026. Treat architecture as risk reduction; pacing, acceptance rate and a consistent device matter more.
Running LinkedIn as your only outbound channel is how a one-week restriction becomes a lost quarter. Read our multi-channel LinkedIn and cold email playbook to see how the two channels back each other up, or talk to our team about a done-for-you outbound system where a restricted account is an inconvenience, not an outage.
By the GenFlows GTM engineering team. We run cold email and LinkedIn outreach for B2B clients and keep every LinkedIn account deliberately under its limits. Quotes attributed to LinkedIn are from its Help Center, User Agreement and Community Report as fetched in September 2026; vendor and community figures are marked directional. Last updated September 2026.