LinkedIn Connection Request: Note or Blank, and How to Test

Add a note to a LinkedIn connection request when you have something real to say. That means a mutual contact, a shared city, an event you both attended, or a business post of theirs you actually read. If you have none of that, send the request blank or with one plain line about who you work with.

If your profile is only a job title and a photo, fix the profile before you send anything. The person you invite will open it before they decide.

The note is one input among several. Who is on your list and what your profile shows usually settle the accept before the note matters much.

My team spent 12 months on LinkedIn outbound for an Australian advertising agency. 1,458 of 3,535 connection requests were accepted, a 41.2% connection rate. That acceptance rate sits well above the published averages further down this page.

No single sentence in an invite produces a gap like that. I see acceptance as the output of the whole setup, and copy cannot rescue a poor list or a sender nobody trusts. How many invites LinkedIn lets you send is a separate question, answered in how LinkedIn limits invites and messages.

Is a note on your invite worth adding at all?

Sometimes. The note is optional, and a blank invite is a perfectly good choice. A note adds one piece of evidence about why your name is in their notifications. It helps only when the evidence is real. A generic line adds nothing a blank invite would not.

For a single request, three things decide whether a note is worth writing:

1. How strong the link between you is. A mutual colleague or a shared event gives the note something to say. A stranger pulled from a filtered search gives it very little.
2. What your profile shows. Most people click through before they accept. A note cannot cover for a page that explains nothing.
3. Who is on your list. A note sent to the wrong person wastes characters, however well it is written.

Whichever way you go, keep the sale out of it. In outreach, the quickest route to being ignored is a pitch in the opening message. The invite only has to win the accept.

The meeting comes later, and even then I sell the call itself, never the service, because a stranger can say yes to a short call long before a contract. The invite asks for nothing yet.

There is also a practical limit. LinkedIn Help states that on a Basic (free) account, personalized notes are capped at 5 invitations, and each note is limited to 200 characters. Premium members get notes on an unlimited number of invitations, with room for up to 300 characters (LinkedIn Help).

On a free account, notes are scarce. Spend them on the people where you have genuine context, and send the rest blank.

When should you send a note, and when should you leave it blank?

Match the invite to how well you know the person and to what your profile shows.

The location row matters more than it looks. When an offer can be sold close to home, I build the first list around one city or region, because a neighbor is easier to start talking to than a stranger on another continent. "I'm in Leeds too" is a short, honest note, and it beats a paragraph of compliments.

The profile rows matter most. A weak note from a strong profile beats a polished note from a ghost. When a visitor finds only a job title and a headshot, the decision takes about 2 seconds, and it goes against you.

The last row is the easiest trap. Someone likes a personal post of yours, you read it as interest, and a few days later you send them a collaboration offer. I would never build an invite on that kind of signal. A like means they saw your post that day, nothing more.

If most of your list has no shared context and you feel tempted to invent some, warm the audience first instead. Warm outbound on LinkedIn explains how that works.

What actually decides whether the invite gets accepted?

Four things sit upstream of the note. With them in place, a plain note is enough.

Fit between the person and your offer. An invite that could go to any of 10,000 people fits none of them well. My first step on any list is the client's current customers, then the companies that look like them. Then I check past projects to see who actually made the buying decision on the client's side. If your list holds the people who really buy what you sell, a short note reads as relevant even when it says very little.

Quality of the list itself. I would take a list pulled from proper databases (Sales Navigator, Apollo, ZoomInfo, UpLead, RocketReach) over a scraped one every time. Strip out vague titles like Assistant, Associate and plain "Manager" before anyone sends anything. Every wrong person who ignores you drags the rate down, and no note fixes that.

What your profile says. I treat the profile as a page written for clients, never a résumé. The headline should say who you help. The About section should say how, with real results.

For a business consulting and coaching firm in the USA, 38 of 98 requests were accepted, a 39% connection rate. On 98 requests the margin is about 10 points either way, so treat it as one data point.

That campaign also ran content marketing alongside the outreach, so the profile was one lever among several. Founder profile work was part of the plan, and I would not start LinkedIn outbound without it.

Industry and seniority. These set your baseline before you write a word. Some markets accept invites freely and others rarely do. A senior buyer flooded with requests judges yours differently from a team lead who gets few. So comparing your acceptance rate with a friend's campaign in another market tells you very little. Compare yourself with yourself, in the same market, over time.

What should the note actually say?

One line, two at most. It should give one reason you are writing to this person and ask for nothing beyond the accept. Templates fail when they read like templates, so short and human beats polished and robotic. Treat the six below as structures to fill with your own words.

Each runs between 106 and 142 characters as written. That leaves room for names inside the 200-character cap on free accounts.

A note on proof. In later messages I put real client results inside the copy, because a number is something the reader can check. In the invite itself, keep proof to the shape of who you work with, as in the peer template. A result with a dollar figure in a 140-character invite reads as a pitch.

The first four templates are where a scarce free-account note earns its place, because the context does the work.

For six marked-up examples, see LinkedIn connection requests that get accepted and answered.

How do you test connection notes without fooling yourself?

Small note tests are easy to misread. Suppose you send version A to 60 people and version B to 60, and get 21 accepts versus 15. That looks like a 10-point win, 35% against 25%. On samples that small, a gap of that size can come from luck alone.

I test constantly, because the more versions you try, the sooner a real winner shows up. The sample sizes below keep that winner honest.

How many invites each version needs. These are standard two-proportion estimates, with a baseline acceptance rate around 30%, 95% confidence and 80% power. In plain words, a real gap of that size shows up about 8 times in 10, and luck alone fakes one less than 1 time in 20:

Turn it around and look at a single rate. At 30% acceptance, 100 invites give a margin of about 9 points either way. 300 invites give about 5 points, and 1,000 give about 3. The Australian campaign above rested on several thousand sends, so its rate is accurate to within about 2 points.

Small samples need the opposite caution. For a software development company in Germany, one of our campaigns shows 52 accepts from 84 requests, a 61.9% connection rate. It is a real number, but on 84 requests the honest margin is roughly 10 points either way. I would not use a sample that size to pick between two notes.

How to run the test.

1. Change only the note. Same sender, same list segment, same weeks.
2. Split one list at random between the versions. Never send version A this month and version B next month.
3. Set the sample size before you start, using the table above, and do not peek and stop early.
4. Track acceptance and replies after acceptance as two separate numbers.
5. When the sample is complete, compare the gap with the margin. If the gap is smaller, call it a tie and keep the cheaper option, which is often the blank invite.

Never compare two reps' notes as a test. Their profiles differ, so you would be measuring the profile, and the note would barely register.

What do the benchmarks say about note versus blank?

SmartReach's 2026 report is a large public dataset on this question. It covers more than 500,000 connection requests sent from January to June 2026, across over 1,800 campaigns run by more than 160 companies (SmartReach). It is the vendor's own campaign data, so treat it as a picture of one platform's customers.

So in this dataset, blank invites were accepted slightly more often, and noted invites drew more replies once accepted. That fits the view I work from: the note is rarely what wins the accept. A real reason in the note may help the conversation that follows, though one platform's averages cannot prove it.

These are still averages across every industry and kind of sender. Your own split test, sized with the table in the previous section, is what answers the note question for your list.

For the wider set of numbers, including what happens after the accept, see LinkedIn outreach benchmarks.

What should never go in the invite note?

• A pitch. No service, no price, no "we help companies like yours grow revenue". A stranger has no basis to judge it, so they don't.
• A link. A link asks a stranger to leave LinkedIn and trust you, before they have even decided to accept.
• A meeting request. The call belongs to a later message, and it still needs selling once you get there.
• Borrowed context. A like on your personal post, a group you both joined years ago, a mutual contact you have never spoken to. If they ask about it, you have nothing to say.
• Your company history. Founding year, headcount and offices are what your profile is for.
• Empty flattery. "Impressive profile" fits every person on LinkedIn, which is exactly why it carries no weight.
• Anything over two lines. If the reason needs a paragraph, it is too weak for an invite.

This is where the invite's job ends. Once someone accepts, the next message has a different task: opening a conversation without wasting the accept. That belongs to what to send after the accept. Keep the two apart. Invite copy that tries to do the follow-up's work turns into a pitch.

If your acceptance rate is stuck, book a call and share your last 100 invites with the notes you used. We can look for what is actually moving the number.