LinkedIn safe sending calculator
find the ceiling before you hit it
Most teams discover their sending limit by crossing it, which costs an account and about three weeks. This moves that discovery to the planning stage. Answer six questions about the profile and the tool returns a weekly invitation figure, the ramp to reach it, and what that produces in a month. Nothing is sent anywhere and nothing is stored.
connection requests a week, which is 15 a day across 5 sending days. This is what one account at your readiness level can carry in a week without pushing into the ceiling. It is a share of the weekly ceiling you entered, not the ceiling itself.
Based on a weekly ceiling of 100, which is your input and a conservative assumption rather than a verified platform fact, at 75% for an account in the Healthy band. The daily figure is derived from the weekly one. LinkedIn reacts to the weekly total, so treat the daily number as a scheduling convenience rather than a rule.
Your seat did not change this number. Sales Navigator and Premium change what you can find and how you can message. They are not a volume upgrade for connection requests, and buying a seat in order to send more is the most common wasted line item in a LinkedIn budget. Change the seat selector and watch the figure above stay exactly where it is. Current Premium and Sales Navigator invitation behaviour is worth checking against LinkedIn's own help centre, and we do not restate it here without a verification date.
Account Readiness Score: 76 out of 100, Healthy
How much room LinkedIn is likely to give this specific account, scored on the five things the platform can see. The score is ours, the rubric is published below, and you can reproduce it by hand.
| Component | Your answer | Points | Available |
|---|---|---|---|
| Account age | 1 to 3 years | +22 | 25 |
| Network size | 500 connections, 500 to 1,499 | +14 | 20 |
| Activity pattern | Weekly, logs in and reads most weeks | +15 | 20 |
| Acceptance rate health | Not measured, neutral middle | +10 | 20 |
| Sending history | Never warned, never restricted | +15 | 15 |
Your ramp to 75 a week
Where you start. Starting at your target volume is the most common and most expensive mistake in this channel. The week count is arithmetic from the two step sizes for your band, assuming nothing goes wrong. It is not a warm up study, because we have not measured one.
| Week | Weekly invitations | Daily equivalent | Check before advancing |
|---|---|---|---|
| 1 | 18 | 3 | Acceptance at or above 30% and no notices, or hold |
| 2 | 30 | 6 | Acceptance at or above 30% and no notices, or hold |
| 3 | 41 | 8 | Acceptance at or above 30% and no notices, or hold |
| 4 | 52 | 10 | Acceptance at or above 30% and no notices, or hold |
| 5 | 63 | 12 | Acceptance at or above 30% and no notices, or hold |
| 6 | 75 | 15 | Target reached, hold here and monitor |
The Hold Rule. Advance to the next step only if both are true at the end of the week. One, your acceptance rate over the last 7 days is at or above 30%, or at or above the account's own trailing 4 week average minus 5 percentage points, whichever is lower. Two, no LinkedIn warning, feature limit or restriction notice arrived during the week. If either fails, hold at the current volume for a full week. If the same check fails a second consecutive week, step down one increment and hold there for two weeks before attempting to advance again. The 30% floor and the 5 point tolerance are our policy, picked conservatively because we absorb the cost of a restricted client account.
What this produces in a month
Set the two rates above and this fills in. The calculator will not guess your acceptance rate or your DM reply rate. There is no industry average preset here because we do not have one, and the ones circulating are published by companies that sell the sending tool. Use your own measured rate if you have one, use the figure we can source as a labelled reference, or model a range. A planning tool that guesses these two on your behalf is a sales tool, because both guesses point the same way and it is upward.
Messaging capacity and your seat
Your DM volume is not governed by a published ceiling the way invitations are. It is governed by how many people accept, which is why the number moves when you change the acceptance rate and not when you change the seat. Set an acceptance rate above and this fills in.
We do not print an InMail credit allowance for the Free seat, because we have not re-verified the current allowances and we are not going to repost a number from an old help article, which is where most of the figures you will find for this come from. We also do not know whether LinkedIn publishes any ceiling on messages to existing 1st degree connections, and we say so rather than implying there is not one.
What would move this most?
- Acceptance rate. Taking it from not measured to 60% and above moves you from Healthy to Established and adds 15 invitations a week.
- Account age is not what is holding this account back. Changing it adds nothing this week.
The address bar holds your inputs, so the link above reproduces exactly this result for anyone you send it to. Nothing about you is stored on our side, because there is no our side to store it on.
Methodology and limitations
What is measured, what is assumed and what is policy. The 59% acceptance and roughly 9% DM reply references are measured, from the control cell of one client programme, and they are offered as labelled references rather than applied as defaults. The weekly invitation ceiling is an assumption you control: it ships conservative and it is not presented as a verified platform figure. Everything else, the five point rubric, the band shares, the ramp step sizes and the 30% floor in the Hold Rule, is LinkedPros operating policy that we chose and are telling you we chose.
It cannot promise you will not get restricted. The ramp is a large reduction in probability, not a guarantee. Accounts that did everything right still occasionally get caught.
It cannot see your account. Everything runs on what you typed. It has no connection to LinkedIn and cannot detect a signal the platform is reacting to that you have not noticed. If this disagrees with what your account is telling you, believe your account.
It cannot model your audience. Acceptance rate is mostly a targeting outcome, and the calculator has no view of who is receiving your requests. A perfectly prepared account sending to the wrong list produces a falling rate that no amount of ramp discipline fixes.
It cannot tell you whether LinkedIn is the right channel. It tells you what LinkedIn can carry. Whether that capacity should be carrying your pipeline target at all starts from the revenue number, not from the channel.
Quick answer
The calculator scores one LinkedIn profile out of 100 on five things the platform can observe about it, places that score in a readiness band, and gives the band a share of the weekly invitation ceiling you are planning against. A mature and active account with a clean history gets most of that ceiling. A three week old profile gets a small fraction of it and a mandatory hold before any increase. The output is a planning figure built from our operating policy, not a limit LinkedIn publishes, and the page shows every step of the arithmetic so you can disagree with a specific number rather than with the result.
How to use it
Answer for one account at a time, and answer about the account rather than about the plan. The most common way to get a useless result is to describe the profile you intend to have in three months. If the account is four weeks old and nobody has logged into it since it was created, say that. The tool is more useful when it tells you something you did not want to hear.
- Account age, network size, activity pattern and sending history describe the profile. These four are facts you can check in under a minute.
- Current acceptance rate is optional. Leave it unset if you have not measured it and the model uses a neutral middle rather than assuming the best.
- The weekly invitation ceiling is an editable input with a conservative default. Replace it with whatever LinkedIn publishes on the day you read this.
- Accounts and sending days scale the result to your actual fleet and your actual working week.
- Planned acceptance rate and DM reply rate are optional and only used for the monthly funnel. Set them as a range if you are guessing, because a range is an honest guess and a point estimate is not.
- A monthly target is optional. Enter one and the page tells you whether your capacity carries it, meets it with nothing spare, or does not reach it.
Run it once per profile. A fleet does not have one number, and the arithmetic that says it does is the single most expensive mistake in multi account LinkedIn. Ten profiles of different ages have ten ceilings, and a campaign that divides a list evenly across them will run the newest one at the rate of the oldest.
What the numbers mean
The Account Readiness Score
A score out of 100 assembled from five components: account age, network size, activity pattern, acceptance rate health and sending history. It is not a measure of how good your outreach is. It is an estimate of how much sending this specific profile can absorb before it starts looking like something the platform should investigate. An old, well connected, genuinely active account with no warnings absorbs a lot. A new, sparse, dormant account absorbs almost nothing, and no tool and no copy changes that.
The band, and the share it gets
The score maps to one of six bands, and each band takes a share of the weekly ceiling you entered. That share is the policy at the centre of the whole model. It is deliberately conservative, because we carry the cost when a client account gets restricted mid campaign, and a rebuild plus the conversation that goes with it is worth far more than the extra invitations we would have gained by running closer to the line.
Safe weekly and safe daily invitations
The weekly figure is the ceiling multiplied by the band share, rounded down. The daily figure is that weekly figure divided across the number of days you told us you send, rounded down. The daily number is a planning convenience rather than a second constraint. What the platform reads is the pattern over time, so an even spread across your sending days is the point of dividing it at all.
The ramp
A week by week schedule from a low starting volume to the target. Weaker bands start lower, step up in smaller increments, and carry a mandatory hold week before the first increase. Each row carries a condition rather than just a number, because a ramp is not a countdown. If acceptance has slipped or a notice has appeared, you hold at the current week instead of stepping up, and the schedule waits for you.
The funnel and the verdict
If you set a planned acceptance rate and a DM reply rate, the page models a month: invitations, accepted connections, messages sent and replies, scaled across your accounts. If you also set a target, the verdict compares the modelled output against it and evaluates on the low end of any range you gave, so a range you entered honestly cannot flatter the answer. When capacity does not carry the target, the page says how many accounts would, because more seats is the only thing that genuinely raises LinkedIn capacity. Copy does not, and neither does a bigger list.
The methodology, and which parts are policy
Three different kinds of number appear on that page and the distinction is the point. Some are your inputs. Some are figures we measured on live campaigns and label as references. The rest is LinkedPros policy, which means a judgement we made and are willing to defend rather than a fact anybody published.
| Element | Kind | Where it comes from |
|---|---|---|
| Weekly invitation ceiling | Your input | Ships with a conservative default because we will not assert a platform figure we have not re-verified. Everything downstream is a share of it, so a wrong ceiling is a correctable error rather than a fabricated fact. |
| Score component weightings | LinkedPros policy | How much we think each observable property should move a sending decision, based on the accounts we operate and rebuild. |
| Band shares | LinkedPros policy | Deliberately below what accounts in each band survive, because we absorb the cost of being wrong and clients do not. |
| Ramp start, step size and hold weeks | LinkedPros policy | Derived from the band. The number of weeks to target is an output of the two step sizes rather than a duration we measured. |
| 59% acceptance reference | Measured | The control cell of one white label programme across advisor workspaces, where the sender read as a peer to the audience. An upper reference for a well matched list, not a median and not a target. |
| Reply rate references | Measured | Roughly 9% on LinkedIn DMs and roughly 1.5% on email, from the same senders against the same list in the same window. Offered as a starting point when you have nothing of your own. |
| Weeks per month | Policy, stated on the page | A monthly figure has to come from somewhere. We use a constant and show it in the workings rather than hiding it in a formula. |
Two hard gates sit in front of the score. A currently restricted account returns nothing except an instruction to stop, because there is no responsible volume for a profile that is already in trouble. A score below the lowest threshold returns a not ready result for the same reason. Neither gate is a soft warning you can click past, and both exist because the number a struggling account most wants is the one it should not be given.
There is also an informational flag for a pattern we see often: a large, old, dormant account. Those look strong on paper and behave badly in practice, because a profile with thousands of connections and no logins for a quarter suddenly sending invitations is a shape the platform has every reason to look at twice.
What it cannot do
It does not know your limit. Nobody outside LinkedIn does. The platform enforces a weekly invitation ceiling that applies broadly and separately applies account quality checks that can restrict a profile well below it, and those checks are not published, not static and not uniform. Anyone printing one confident daily number is repeating folklore, usually from an old help article, usually with something to sell.
- It cannot see your account. Every input is you describing the profile, and a hopeful description produces a hopeful number.
- It does not model your audience. A perfectly safe volume aimed at the wrong list still fails, just without a restriction to explain it.
- It does not model connection request notes, InMail credits or content activity, all of which interact with acceptance and none of which we have measured well enough to put a coefficient on.
- The band shares are our judgement. A team with a higher tolerance for losing an account would reasonably set them higher, and the workings panel exists so you can see exactly what you would be changing.
- Platform limits move. The ceiling field is editable precisely because this page will eventually be out of date and we would rather it be correctable than authoritative.
- The funnel multiplies rates you supplied. If those rates are optimistic, the output is optimistic with more decimal places, which is the classic way a spreadsheet launders a guess into a plan.
The honest summary: this is a planning instrument, not a permission slip. It exists to stop teams typing a confident number into a limit field in week one, which is the most common cause of a lost account we see. Once you are sending, the account itself is the better instrument, and acceptance rate is the signal to read first because it moves before anything else does.
Capacity is one half. The audit works on the segment and the offer, which is the half a calculator cannot reach.
Why we publish a conservative number
Almost every figure about LinkedIn sending limits on the open web is published by a company selling LinkedIn automation seats, and their commercial interest points at a bigger number. Ours points the other way. LinkedPros is part of the Outbound Pros group and the group runs LinkedIn outreach as a managed service, which means when an account is restricted mid campaign we carry the rebuild and the client conversation. That is the bias in this tool and it is the one we would rather you had. If you want to see how the same discipline reads applied to software, the tool reviews and the head to head comparisons use the same rule about what we will and will not assert.
Questions about the calculator
Is the number it gives me the LinkedIn limit?
No, and the distinction matters. LinkedIn enforces a weekly invitation ceiling plus account quality checks that are not published. This tool takes the ceiling as your input and then applies our own conservative share on top of it based on how much sending the profile looks able to absorb. The result is a planning figure that sits under the platform limit on purpose.
Why does it ask for the weekly ceiling instead of telling me?
Because we have not re-verified that figure recently enough to assert it, and a planning tool that invents a platform fact is worse than one that asks. The default is deliberately conservative. Replace it with whatever LinkedIn publishes on the day you read this, and every output adjusts as a share of what you entered.
Does my data go anywhere?
No. The whole thing computes in your browser. There is no signup, no email capture, no analytics event carrying your answers and no request to a server with your inputs in it. Your settings are readable in the page URL, which is how you share or bookmark a scenario, so avoid pasting a link somewhere public if the account details in it are sensitive.
My tool lets me send more than this. Who is right?
Both, in the sense that your tool is reporting what it will execute and this is reporting what we think the account can carry. Automation platforms send whatever number you type in. That is not a safety judgement, it is a text field. The gap between the two figures is the risk you are choosing to take, and the point of the page is to make it visible rather than to forbid it.
How should I use it across a fleet?
One profile at a time, and then add the results up. Do not average the fleet and apply the average, because the profile that gets restricted is always the newest one being run at the pace of the oldest. Set the accounts field to model total output once you know each profile individually, and keep the per account ceilings in whatever tool you are sending from.