LinkedIn connection message templates that get accepted (2026)

Write your connection note to 200 characters, not 300. LinkedIn's help documentation gives 200 as the limit for a basic account and reserves 300 for Premium, so a template library built at 300 is wrong for most of the people using it. Then sort your notes by the evidence they cite rather than by the persona they target, because the note's only job is to make the shared context visible in one glance. Everything below is written to that budget, with real character counts.

The character budget is 200, and LinkedIn's own pages disagree on the rest

Start with what is published, because most advice on this topic is written against a limit that does not apply to the reader.

LinkedIn's help page on personalising invitations to connect states that "Each message can be up to 200 characters" and that "You can add a personalized message to up to three connection requests per month" on a basic account, with Premium members able to "send unlimited personalized invitations to all their connection requests (connection request limits may still apply)".

LinkedIn's separate help page comparing InMail messages with personalised invitations describes basic accounts as getting "up to five personalized invitations to connect with a character limit of 200", and Premium accounts an "unlimited number of personalized notes with their invitation/connection requests each with a character limit of 300".

Two of LinkedIn's own pages, two different monthly figures — three and five — and one consistent character limit of 200 for basic accounts. That is worth stating plainly rather than resolving artificially: the character limit is documented and stable at 200; the monthly personalised-note allowance is not, and you should check the behaviour on your own account rather than plan around either number.

The practical drafting consequence is simple. Write to 200. A note that fits 200 characters fits every account type; a note written to 300 is unusable on the most common one, and a tool that silently truncates it will cut your sentence in the middle.

Acceptance rate is the resource you are actually spending

The note is not free, and understanding what it costs changes how you write it.

LinkedIn's help page on the types of restriction for sending invitations names three triggers in its own words: "You've sent many invitations within a short amount of time"; "Many of your invitations have been ignored, left pending, or marked as spam by the recipients"; and "If you send an excessive number of invitations and we suspect the use of an automation tool, we may suspend or restrict your account."

Read the second one again. An ignored invitation costs you something even though nothing happened. It sits pending, it drags the ratio LinkedIn is watching, and the pending pile is itself part of the picture. There is no email equivalent of this: an unopened email is close to free, while an unaccepted invitation is a small, permanent debit.

That gives the note a precise job. It is not there to persuade, sell, or introduce. It is there to raise the probability of a decision — yes or no — over the probability of a shrug. A note that produces a fast no is worth more than one that produces a pending request, because the no clears and the pending does not.

Two related numbers are worth tracking weekly per sender: trailing acceptance rate and pending count. Both degrade before a restriction, and neither is visible in a send counter. Autocloz's LinkedIn pacing acts on the first of them directly — when the trailing accept rate on a sender falls below 30% with a sample of at least ten invitations, it halves that sender's effective daily ceiling for the dispatch rather than continuing at full speed into a falling signal.

Eight notes, sorted by the evidence they cite

Sorting templates by persona produces interchangeable copy with a swapped job title. Sorting by evidence type produces notes that could only have been written to that person, which is the entire point. Every example below is a real character count of the text as written.

Evidence they created

Something they published — 149 characters. "Read your post on why your team dropped attribution reporting. We argued the same thing last year and lost. Would like to follow how it goes for you."

Why it works: it names a specific claim, adds one line of your own experience that is not flattering to you, and asks for nothing. The disagreement-or-shared-failure move is underused and it reads as human because a template would not admit to losing.

Something they said in public — 110 characters. "Caught your conference session on onboarding debt - the point about the 90-day cliff landed. Happy to connect."

Why it works: it is short enough to read in the notification preview, and the specific detail proves attendance. Vagueness here is fatal — "loved your talk" is indistinguishable from having read the title.

Evidence visible on the profile

Role and scale — 143 characters. "You run RevOps at a 40-person team and I build outbound tooling for teams that size. No pitch - would just like to be in each other's networks."

Why it works: it states the relevance and then explicitly disclaims the ask. "No pitch" is a promise you then have to keep, which is the point.

A role change — 141 characters. "Congratulations on the move to Head of Sales. I work with teams making exactly that jump and would like to follow how the first quarter goes."

Why it works: a role change is a genuine trigger and a socially normal reason to reach out. It also dates the note, so it has to be sent within a few weeks or it reads as scraped.

Evidence from a shared context

A shared connection — 136 characters. "We are both connected to the Bangalore RevOps group and to two of the same people. I work on outbound tooling and would like to connect."

Why it works: the overlap is checkable in one click, which is what makes it credible. Naming a specific mutual is stronger still, though only where you genuinely know that person.

A shared room — 136 characters. "We were both at the SaaS founders meetup last month. I work on multichannel outbound and would like to keep the conversation going here."

Why it works: co-presence is the cheapest form of trust there is, and it expires. Send it inside a fortnight.

A shared background — 112 characters. "Fellow alum working in RevOps tooling. Would like to connect and follow what you are building on the sales side."

Why it works: it is short, true, and asks nothing. Alumni notes have a low ceiling and a very high floor.

Evidence about the company

A public hiring signal — 132 characters. "Your job post for a second SDR is what caught my eye - we work with teams at exactly that hiring point. Happy to connect either way."

Why it works: it names the signal, which proves you looked, and "either way" removes the implied obligation. This is the closest any of these gets to commercial, and it still does not pitch.

Notice what none of them contain: a link, a calendar, a question that requires research to answer, a compliment about the company, or the phrase "I would love to". Each one is one specific observation plus one low-cost ask, and each one is written to fit 200 characters with room for a longer name.

Autocloz's free plan runs LinkedIn invitations, messages and profile actions as steps in a sequence alongside email, calling, SMS and WhatsApp on 5 users and 10 mailboxes — start free and set a per-sender daily ceiling before you write a single note.

Why the blank invitation is a legitimate control rather than a lazy option

The blank connection request deserves better than its reputation, and the argument for it is structural.

A note supplies context that is not otherwise visible. Where the context is visible — you are in the same industry, the same city, obviously adjacent roles, with mutual connections shown on the request itself — the note adds nothing and introduces a risk: it gives the recipient a reason to categorise you. A note that reads even faintly like a pitch converts an easy yes into a considered no.

So the honest rule is a decision, not a preference:

• Context visible from your headline and mutuals — send blank. The profile is doing the work.
• Context not visible — send a note, because there is no other way to supply it.
• Context visible but unflattering — you are an obvious vendor to an obvious buyer — send a note that names it. Pretending otherwise reads worse than saying it.

Treat blank-versus-note as an experiment you actually run on comparable segments rather than a position you hold. It is one of the few LinkedIn variables you can test cleanly, because the two arms differ in exactly one thing.

What happens after acceptance, and how long to wait

The most common failure is not the note. It is the message that lands four minutes after acceptance.

An instant follow-up is the clearest automation signal available to a recipient, because a human who just accepted a request does not have a pitch ready. It also wastes the acceptance: the person said yes to being connected, not to a conversation, and the two are separate decisions taken minutes apart only by software.

A workable pattern: let the acceptance sit. Engage with something they post if there is something worth engaging with. Then send a first message that references why you connected, in one or two sentences, and asks a question that costs them under thirty seconds to answer. The difference between spending an InMail credit and earning a connection first is worth reading before deciding which of the two you are actually running.

Autocloz sequences the invitation and the follow-up message as separate steps with their own daily caps, adds a random 60 to 300 seconds on top of the sender's configured minimum gap after each action, and skips the LinkedIn step entirely when the lead has replied on any channel in the trailing fourteen days. That last rule matters more than it sounds: a scheduled invitation landing on somebody who answered your email on Tuesday is the single clearest way to tell them the whole thing is automated.

Withdrawing, and the three-week rule almost nobody plans for

Pending invitations are a liability, so withdrawing old ones looks obviously correct. There is a catch that changes the calculation.

LinkedIn's help documentation on withdrawing an invitation states that after you withdraw one, you will not be able to resend an invite to the same recipient for up to three weeks. It also states that the recipient is not notified, that withdrawing stops further reminder emails reaching them, and that there is no option to bulk withdraw invitations sent to multiple recipients.

Three consequences:

• Withdrawal is a per-person decision with a cooling-off period, so withdrawing an invitation to somebody you actually want costs you a month of access to them.
• It is manual. No bulk action exists, so a large pending pile is a genuine time cost, not a button.
• It does not lift a restriction. LinkedIn's page on reaching the invitation limit says directly that "Withdrawing pending invitations will not remove the restriction."

The right response is to send fewer, better-targeted invitations rather than to plan a monthly withdrawal sweep. Autocloz's stale-invite handling marks the ledger after a configurable age so the acceptance-rate and pending-count arithmetic stays accurate, and it deliberately does not attempt the withdrawal on LinkedIn itself.

How to test a note honestly when the sample is small

Acceptance-rate testing is where most LinkedIn advice quietly becomes fiction, because the samples involved are far too small for the confidence people assign to them.

Four rules that keep it honest:

1. Change one variable. Note versus blank, or evidence type A versus evidence type B. Not both, and not the targeting at the same time.
2. Hold the segment constant. Acceptance rate varies far more by seniority, industry and graph distance than by wording, so a test across two different lists measures the lists.
3. Wait for the pending to resolve. Acceptance is not instant; measuring at 48 hours systematically favours whichever arm went to people who happened to be online.
4. Do not read a difference of a few points as a result. At the volumes a single account can send, small differences are noise. What you are looking for is a difference large enough to be obvious.

The variable that moves acceptance most is not in the note at all — it is who you sent it to. A precise list beats a clever sentence every time, and building a targeted list before writing anything is the higher-leverage half of this work. If you are weighing LinkedIn-first tooling for it, the trade-offs against Waalaxy set out where a dedicated LinkedIn tool goes deeper and where a multi-channel system does.

What a good note cannot fix, and what Autocloz does not do

A note cannot make a badly chosen recipient a good one. If a third of the people on your list have no plausible reason to accept, no wording rescues the ratio, and the ratio is what LinkedIn is measuring. Targeting is the lever; the note is a multiplier on it.

A note also cannot make LinkedIn automation contractually safe. Section 8.2 of LinkedIn's User Agreement tells members not to "use bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages, create, comment on, like, share, or re-share posts". Every tool in this category operates in that grey zone, and pacing lowers the probability of a restriction without removing it. If losing a specific profile would be unacceptable, send that profile's invitations by hand.

Three product boundaries stated plainly. Autocloz trims an invitation note at 300 characters before dispatch, because that is LinkedIn's ceiling for a Premium account and the higher of the two documented limits. If the connected account is a basic one, LinkedIn's own documented limit is 200 — so a 260-character note can pass the product's trim and still be refused by LinkedIn. Write to 200 and the question never arises.

Autocloz does not supply LinkedIn profile data or contact data of any kind, so list quality stays entirely yours. It does not withdraw invitations on LinkedIn — the stale-invite action updates the internal ledger so pacing arithmetic stays honest, and the withdrawal itself remains a manual job inside LinkedIn. And sequencing invitations, messages and profile actions with their own per-kind daily caps reduces the risk of a restriction rather than eliminating it, which is why the shared suppression list and gate across all five channels matters more than any single channel's pacing: the complaint you avoid is worth more than the invitation you send.