LinkedIn Connection Acceptance Rate: When to Add a Note and How to Write It

Should you add a note to your LinkedIn connection request? It's one of the most argued-about questions in B2B prospecting, and most of the answers online come with a confident percentage attached.

Here's the short version: a note barely moves your acceptance rate, and it can move what happens afterward quite a bit. The public studies that compared the two options found similar acceptance with and without a note, and more replies on the note side. But there are only a few of those studies, they come from outreach vendors, and they measure slightly different things. So this guide won't hand you an "ideal acceptance rate" or promise percentage points. It shows you how to decide, case by case, whether a note is worth it, and how to write one that fits the limits LinkedIn documents today.

The detailed comparison of sources, including what each one measured and why they disagree, is further down, in the section on why the studies disagree. The main focus here is the decision and the writing.

What is connection acceptance rate, and why does it matter less than you think?

Acceptance rate is the share of connection requests that get accepted. Send 100, get 30 accepted, and your rate is 30%.

It matters because on LinkedIn there's no direct message without a connection (outside of InMail). But it isn't the KPI that pays the bills. A high acceptance rate with no conversations is worth less than a lower rate that starts real ones. Acceptance gets you in the door; what happens next decides whether prospecting works.

Acceptance rate is still worth watching because:

It drives cost per conversation. The lower it is, the more requests you need for the same number of conversations, and LinkedIn caps how many invitations you can send.
It's a fit signal. A rate that stays low usually points to your profile, your targeting, or your list before it points to your note.

Does a note increase acceptance rate?

Based on the public data, barely, if at all. In the largest published dataset, from the agency Belkins using data from the outreach platform Expandi (11.5 million connection requests sent in 2025), requests with a personalized note were accepted 25.3% of the time and requests without one 27.6% of the time. In a smaller test by AiSDR, with about 450 leads per group, acceptance was 34.6% with a note and 37.3% without.

Two sources, same direction: blank requests get accepted slightly more often, and the gap is small. Where the note pulls ahead is replies. Belkins measured 8.2% vs. 5.3% (replies over messages sent); AiSDR measured 19.2% vs. 2.6% (replies over leads contacted). Those two figures use different denominators and can't be compared with each other, but they point the same way. Belkins' own read is that a note acts as a "soft filter": people who accept knowing what the outreach is about have already opted into a conversation.

What that means for your decision:

If your bottleneck is connection volume, the note isn't your lever. Profile, targeting, and list quality matter more.
If your bottleneck is conversations after acceptance, a good note tends to help, as long as it's specific.
Neither study separated good notes from bad ones. The claim that a generic note is worse than no note is a reasonable hypothesis, not a published measurement.

For more on why acceptance depends on who's sending more than on the text, see the B2B prospecting benchmarks for 2026 and read their numbers with the same caution.

Why do the studies' numbers disagree so much?

Because each study uses a different denominator for replies, the samples look nothing alike, the publishers sell prospecting products and the same study changes between versions. Belkins' "8.2% vs. 5.3%" and AiSDR's "19.2% vs. 2.6%" cannot be lined up side by side.

1. Each study uses a different denominator. Belkins calculates reply rate as replies ÷ messages sent. Expandi calculates the connection-note reply rate as replies ÷ requests sent. AiSDR doesn't define it in writing, but its own numbers show the rate is over leads contacted: a 19.2% response rate on 34.6% acceptance is the "roughly 55% of accepted connections" the article cites. Three denominators, three numbers that don't compare.

2. Sample size and selection. Eleven and a half million requests mix every kind of campaign and note. Nine hundred leads picked from intent signals and warmed up before the request are a different universe. A small, well-designed test shows bigger effects; a giant aggregate dilutes everything.

3. The publishers sell the product. AiSDR sells an AI SDR that writes notes; Belkins is a demand-generation agency; Expandi is an automation tool. That doesn't invalidate the data, but it is a reason to read the study design before quoting the number. The same goes for this article: Chattie also sells LinkedIn prospecting.

4. The same study changes between versions. Belkins' page was published in July 2025 and updated in June 2026, with 11.5 million requests and 25.3% vs. 27.6% acceptance. An earlier version, still quoted by Expandi in its connection message templates guide, cites "20M+" attempts and 26.42% vs. 26.37%. If you quote it, open the page and cite what is there.

What does no public study answer?

Five questions still have no public measurement: the effect by seniority, the effect of note length on connection requests, the effect by market, the effect on account health and the effect on mobile. Anyone asserting one of them is offering an opinion.

Effect by title or seniority. Claims like "executives respond more to personalization" have no public note-vs-no-note comparison behind them.
Effect of note length on connection requests. What exists is about InMail: in LinkedIn's study, InMails under 400 characters responded 22% above average. Connection requests have a different limit and a different context.
Effect by market. Neither comparison breaks results out by country or language.
Effect on account health. LinkedIn doesn't publish restriction criteria or a minimum acceptance rate.
Effect of the note on mobile. The idea that a note adds friction in the app is a hypothesis; there is no public measurement.

What are LinkedIn's current note limits?

LinkedIn has changed these rules in recent years, and plenty of content still repeats old numbers. Here's what LinkedIn Help says as of October 2026:

What	What LinkedIn documents	Page
Note length (free account)	"Each message can be up to 200 characters"	Personalize invitations to connect
Number of notes (free account)	A personalized message on "up to three connection requests per month"	Personalize invitations to connect
Premium	Unlimited personalized invitations ("connection request limits may still apply")	Personalize invitations to connect
Note length (Premium)	Up to 300 characters	Differences between InMail Messages and personalized invitations

One honest caveat: LinkedIn's own pages don't match. The main page on personalized invitations says three notes a month for free accounts; the page comparing InMail and invitations still says "up to five." When in doubt, check what your account shows when you click "Add a note."

What this means in practice:

On a free account, notes are scarce. Three (or five) a month can't support any volume strategy. Save them for people you have a real reason to reach.
Write for 200 characters, even on Premium. If your note doesn't fit in 200, it's probably trying to do too much.
LinkedIn also limits invitations overall, not just notes, and its help page doesn't publish the exact number. Don't treat "100 a week" or "200 a week" from a blog post as an official rule.

When to add a note

Add a note when you have something that only makes sense for that person. In practice, four situations:

1. They engaged with you or your content. They commented on your post, reacted to something your team shared, or were at an event you attended. A note that references that interaction is expected; skipping it feels odd.

Example: "Saw your comment on [name]'s post about [topic] — I see it differently on [point]. Would be good to connect."

2. There's verifiable shared context. A mutual connection they actually know, the same community, the same event. Mention the context, not just the name.

Example: "[Mutual connection] mentioned you're building out [team or project]. We worked together on [context]. Good to connect."

3. There's a concrete trigger on their profile or at their company. A recent role change, a post on a topic you know well, an expansion the company announced. The reference has to be specific enough that they can tell you looked.

4. It's a strategic account. For your few highest-value prospects, five minutes reading their profile and activity before writing is time well spent. That's also where a free account's scarce notes pay off most.

When to send it blank

You'd write the same thing to everyone. The test is simple: could this note go to anyone with the same title if you swapped the name? If so, don't send it.
You have no trigger for that person. A clean request from a clear profile that fits your audience is a safe default.
The note would be a pitch. Product, service, link, meeting request: all of that waits until after they accept.
You're at volume with no time to research. The first message after acceptance is where you personalize.

How to write a note that works

There's no public study measuring which note structure converts best for sales connection requests. What follows comes from the format itself (200 characters, read in seconds, often on a phone) and from what's known about short, one-to-one messages.

The closest data point comes from a different LinkedIn product. In a study of tens of millions of recruiter InMails, LinkedIn found that InMails sent individually got response rates roughly 15% higher than bulk sends, and that the shortest ones (under 400 characters) responded 22% above average. That's recruiter-to-candidate InMail, not B2B connection requests, but the direction fits: short and one-to-one.

The structure:

[Specific trigger] + [One line on why you want to connect]

No pitch, no link, no question they have to answer before accepting.

Templates to adapt (all under 200 characters):

Recent post: "Read your post on [topic] — the point about [detail] matches what I'm seeing in [context]. Would like to follow your work here."
Role change: "Saw you just took over [role] at [company]. I work with [function] teams going through that kind of transition. Good to connect."
Event: "We were both at [event]. Enjoyed the discussion on [topic] — connecting to keep it going."

What doesn't count as personalization

Name and title dropped into a template, nothing else
A vague industry reference ("I see you work in tech")
Generic praise ("your profile is impressive")
City or school, which any tool can pull in bulk
The company name with no fact about the company

For more connection message templates, with and without a note, see 12 real B2B connection message examples.

How to personalize at scale without it becoming a template

Personalization takes time, and prospecting takes volume. The way out is not to personalize everything the same way:

Use three tiers. Most requests go out blank. A middle group gets a segment-level note with a specific reference to what that kind of company is going through (for example, companies that just opened sales roles). Strategic accounts get a note written for the person.

Write by trigger, not by title. "VP of Sales" isn't a trigger. "VP of Sales who just started" is.

Use the first message after acceptance. If the request went out blank, that's where the context goes. See how to write LinkedIn messages that don't sound salesy.

If AI writes your notes, have it find the trigger, not fill the template. A grammatically perfect note with no real observation is just a better-written template. Hand-reviewing notes to strategic accounts is still worth it.

What drives acceptance beyond the note?

If the note barely moves acceptance, what does? No public study ranks these factors, so treat this as reasoning, not measurement:

Profile credibility. A clear photo, a headline that says what you do, and experience consistent with the people you're reaching. The prospect decides in seconds whether you look legitimate.
Targeting. Reaching outside your ICP depresses acceptance regardless of the note.
Shared network. Mutual connections give the prospect a reason to trust the request. Filtering for them in Sales Navigator is a common approach.
Recent activity. People who are active on LinkedIn are the ones who see and act on requests.
Account history. Newer accounts should ramp up gradually; see LinkedIn Automation: What Is Allowed and What Can Get Your Account Banned.

To pull these together into a working process, see How to Prospect on LinkedIn B2B.

Does a note affect account health?

LinkedIn doesn't publish the criteria it uses to restrict invitations, or any minimum acceptance rate. Any number you read along those lines (for example, "below 20% triggers restrictions") is the writer's guess, not a platform rule.

What's reasonable to assume, as a hypothesis: lots of ignored or declined requests don't help your account, and poorly targeted requests get declined far more than any note will cause. That's why targeting and volume matter more than the note decision.

How to measure it in your own operation

The only way to know whether a note helps with your audience is to measure. The basics:

1. Track two rates separately. Acceptance rate (accepted ÷ sent) and post-acceptance reply rate (connections that replied ÷ connections accepted). The metric that combines them is conversations per 100 requests: acceptance × reply rate × 100.

2. Do the math before you conclude. A hypothetical example: with a note, 25% acceptance and 8% replies gives you 2 conversations per 100 requests; without one, 28% and 5% gives you 1.4. The note loses on acceptance and wins on the total. Plug in your own numbers. The guide to raising your LinkedIn reply rate covers the post-acceptance side.

3. Don't trust small samples. Using the standard two-proportion calculation (95% confidence, 80% power), detecting a drop in acceptance from 30% to 25% takes about 1,250 requests per group. Small differences over a few dozen requests are noise. That's arithmetic, not market data.

4. Change only the note. Same account, same ICP, same first message after acceptance, same period. If the note and the sequence change together, you can't tell which one made the difference.

5. Randomize who goes into each group. Shuffle the list and assign leads at random. If the "best" leads get a note and the rest get a blank request, you are testing the list, not the note.

6. Fix a reply window. Count replies up to a fixed number of days after acceptance (14 days is a practical window), the same for both groups.

7. Decide on the total, including a tie. If conversations per 100 requests come out similar, the choice becomes one of effort: blank requests take less time. If the note wins clearly, it pays for itself. Measuring replies, not just acceptance, needs a bigger sample: detecting 5% vs. 10% replies among accepted connections takes about 435 accepted connections per group, which at 30% acceptance is roughly 1,450 requests per group (hypothetical example, same calculation as item 3).

Where Chattie fits

Chattie doesn't publish platform usage data, so this guide doesn't include any "Chattie rate." What we can describe is how the product works, because that's configuration:

Fixed daily caps: 25 connection requests, 40 conversations, and 200 LinkedIn actions per day, sent only during business hours. These caps are fixed by default; above 25 connection requests per day, only with the user's own risk consent. They're what keep account risk low (never zero).
Pending invitations withdrawn automatically after 15 days.
Three sending modes: in Autopilot the AI sends the message; in Semi-pilot the AI sends until the lead replies, after which a human approves each response; in Copilot the AI writes and a person approves before it goes out. Copilot is the natural fit for hand-reviewing messages to strategic accounts.

At 25 requests a day, getting to 1,250 requests per group in a note test takes a while on a single account. That's one more reason to make the note decision with the logic in this guide and use a test to confirm it, not to discover it from scratch. Learn more at Chattie.

FAQ

Common questions about connection notes and acceptance rate for B2B sellers.

Does adding a note to a LinkedIn connection request increase acceptance rate?

The public data says no. Belkins' analysis of 11.5 million requests sent in 2025 found 25.3% acceptance with a note and 27.6% without. A smaller AiSDR test found 34.6% with a note and 37.3% without. Where notes pull ahead is reply rate, not acceptance.

What is LinkedIn's character limit for connection notes?

According to LinkedIn Help, 200 characters on a free account. One of LinkedIn's help pages lists 300 characters for Premium members. Writing for 200 is a good habit on any account.

How many notes can I send per month on a free account?

LinkedIn's main help page says free accounts can add a personalized message to up to three connection requests per month; another help page still says five. Premium accounts get unlimited notes but are still subject to overall invitation limits.

Is there an official benchmark for a good acceptance rate?

No. LinkedIn doesn't publish one. As a reference point, Expandi's 2026 benchmarks report a 28.5% average acceptance rate across 13.2 million requests on its platform, but that's one vendor's user base, not a target for your account.

Can a connection note cause account restrictions?

LinkedIn doesn't publish its restriction criteria or any minimum acceptance rate. It's reasonable to assume that lots of declined requests don't help, but targeting and volume matter far more than whether you add a note.

What counts as a genuinely personalized note?

A note that only makes sense for that person: it references a post, an interaction, a mutual connection they actually know, or a concrete fact about their company. Name and title in a template don't count.

Should I add a note to every request?

No. Add one when you have a specific trigger for that person, and send it blank when you don't. A note that would work for anyone in your audience adds nothing.

Why do notes seem to increase replies?

Belkins' explanation is selection: people who accept a request with a note already know what it's about, so they arrive more willing to talk. That doesn't prove the note causes the reply, but it's consistent with both public studies that compared the two options.

Which matters more, acceptance rate or reply rate?

Neither on its own. Track conversations per 100 requests, which combines both. A lower acceptance rate with more replies can beat a higher acceptance rate with few.

Conclusion

The note-or-no-note question carries less weight for acceptance than most advice suggests: public studies find small differences, with a slight edge for blank requests. What changes is the conversation after acceptance, where notes come out ahead.

The operating rule is simple. No trigger, send it blank. Real trigger, write a short, specific note with no pitch, within 200 characters. On a free account, save your few notes for the people who merit them. And track conversations per 100 requests, not just acceptance.

References

LinkedIn Help — Personalize invitations to connect: 200-character limit and notes on up to three requests per month on free accounts; unlimited on Premium (linkedin.com/help/linkedin/answer/a563153)
LinkedIn Help — Differences between InMail Messages and personalized invitations: up to five personalized invitations on free accounts (200 characters) and 300 characters on Premium (linkedin.com/help/linkedin/answer/a6239760)
Belkins — LinkedIn outreach study (Michael Maximoff, updated June 2026): 11.5 million requests from 2025 via Expandi; 25.3% acceptance and 8.2% reply rate with a note, 27.6% and 5.3% without (belkins.io/blog/linkedin-outreach-study)
AiSDR — connection note test (August 2026): about 450 leads per group; 34.6% acceptance with a note vs. 37.3% without; 19.2% vs. 2.6% response (aisdr.com/blog/linkedin-connection-request-results)
Expandi — LinkedIn Outreach Benchmarks 2026: 13,218,869 requests from 13,302 accounts; 28.5% average acceptance (expandi.io/blog/linkedin-outreach-benchmarks-2026)
Expandi, LinkedIn connection message templates: quotes the earlier version of the Belkins study, 26.42% vs. 26.37% across 20M+ attempts (expandi.io/blog/linkedin-connection-message-templates)
LinkedIn Talent Blog — These InMails get the best response rates (2022): individually sent InMails respond about 15% better than bulk; the shortest, 22% above average. Measures recruiter InMail, not sales connection requests (linkedin.com/business/talent/blog)