Asset 20 8 2
Does AI recommend your business? Run the free check →

Join 15,000 business owners, marketers and entrepreneurs. The Sunday newsletter you'll be annoyed only arrives once a week.

Article

What Should a LinkedIn Cold Message Say?

If you are skim reading
The short version: a LinkedIn cold message that gets replies says almost nothing about you and a lot about a specific problem the other person has, fits inside the first two lines before the "see more" cut-off, and asks for one small thing rather than a call.

The short version: a LinkedIn cold message that gets replies says almost nothing about you and a lot about a specific problem the other person has, fits inside the first two lines before the "see more" cut-off, and asks for one small thing rather than a call. Templates help, but the structure matters more than the wording, and most people copy the wording and skip the structure.

Why most cold messages get deleted in under three seconds

LinkedIn cuts off a message preview after roughly the first 100 to 140 characters on desktop, and even less on mobile. That means your opening line is doing all the work. If it starts with "I hope this message finds you well" or "I came across your profile and was impressed," you have used your entire preview window on nothing. The reader never taps to expand it.

A connection request note has an even harder limit: 300 characters, full stop. That is about two short sentences. You do not have room for your company history, your mission statement, or three bullet points about what you do. You have room for one sentence that proves you know who they are and one sentence that makes a small, specific ask.

Here is the uncomfortable part most templates skip over: personalisation that is obviously inserted by a tool is worse than no personalisation at all. "I saw you work at {{company}} as {{job_title}}" reads exactly like what it is, a merge field. People have become very good at spotting that pattern in the last two years, partly because so many automation tools make it the default. If you are going to mention their company or role, say something about it that a scraper could not have pulled from a data field, something that shows you looked at the post, the comment, or the job change they announced.

The structure that gets replies (step by step)

There are five pieces, in this order, and the whole message should be under 500 characters for a first touch:

  • 1. A specific, verifiable observation about them, their company, or something they posted in the last 30 days. Not a compliment, an observation.
  • 2. One sentence of context on why you are messaging, tied to that observation, not to your product.
  • 3. A relevant proof point or credibility signal, one line, no bullet list, no case study link.
  • 4. A low-friction ask. Not "can we jump on a 15-minute call," but "worth a quick look?" or "want me to send the short version over?"
  • 5. An easy out. "No pressure if it's not relevant right now" sounds like it weakens the message. It raises reply rates because it removes the social cost of saying no, which is often the actual reason people stay silent instead of replying.

Notice what is missing: your job title, your company's 20-year history, and a list of client logos. All of that belongs on your profile, not in the first message.

A worked example

Say you run a small bookkeeping firm that works with online retailers doing between 200,000 and 2 million pounds a year in revenue. You want to message e-commerce founders on LinkedIn. A weak version looks like this:

"Hi [Name], I hope you're doing well. I run a bookkeeping firm that specialises in helping e-commerce businesses stay on top of their finances. We'd love to help you too. Do you have 15 minutes this week for a quick call to discuss?"

Nobody replies to that because it could have been sent to 10,000 people unchanged, and the founder knows it. Here is the same offer, rebuilt with the five pieces:

"Hi [Name], saw your post about moving stock off Shopify's native reports into a real dashboard, that's a jump most founders put off too long. I work with retailers around your size on exactly that gap between the sales numbers and the actual cash position. One client found they'd been under-reporting VAT by about 4,000 pounds a quarter before we caught it. Want me to send over the three questions we always check first? No pressure if finance is sorted already."

Same offer, same company, completely different message. It references something real and recent, states a number, and asks for something that takes thirty seconds to say yes to.

Three templates you can adapt today

Template 1: The connection request (300-character limit)

"Hi [Name], saw you're hiring for [role] again this quarter, that usually means growth is outpacing process. I help [niche] with [specific outcome]. Would be good to connect either way."

Template 2: The first message after they accept

"Thanks for connecting, [Name]. Quick one: I noticed [specific, recent, verifiable detail]. We've helped a few similar businesses with [specific outcome, with a number if you have one]. If it's useful I can send the one-page version, no call needed. If not, no worries either way, good luck with [specific thing they mentioned]."

Work with me

Want AI doing the heavy lifting in your marketing?

I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.

Template 3: The re-engagement message, 10 to 14 days later, if there was no reply

"Hi [Name], not chasing, just flagging this in case it got buried, I know inboxes get busy. If the timing's off that's completely fine, happy to check back in a few months instead."

That last line matters more than it looks. Giving someone permission to say "not now" rather than forcing a yes or no on "not interested" keeps the door open and often brings a reply weeks later when their situation changes.

What to leave out, even if it feels like it is helping

  • Attachments or links in the first message. Many inboxes flag or deprioritise messages with links, and it signals "sales pitch" before they have read a word.
  • Three or more questions in one message. One question gets answered. Three questions get ignored because answering feels like work.
  • Emojis in a B2B first message, unless your audience uses them first. It is a small thing but it shifts tone from professional peer to marketer, and the switch happens in under a second of reading.
  • "Just following up" as an opener. It tells the reader you have nothing new to say, which is true most of the time, so say something new instead, even something small.

How many messages, and how often

LinkedIn's own published benchmarks have put InMail response rates somewhere in the 10 to 25 percent range depending on industry and how targeted the list is, compared with connection request acceptance rates that typically sit much lower when the note is generic. The gap between those two numbers is entirely explained by targeting and message quality, not by some clever trick. A list of 50 well-researched prospects with a tight message will outperform a list of 2,000 scraped contacts with a merge-field template every time, and it takes less of your week to run.

If you are sending connection requests with a note at any volume, it is worth knowing LinkedIn's own weekly caps before you plan a campaign around a number that will get your account restricted. I cover the actual limits and how they have shifted in how many LinkedIn connection requests you can send per week, and the practical side of writing a note that does not get ignored in how to send LinkedIn auto connect requests with a personal note.

Should you automate any of this?

Automation can handle the sending and sequencing, it cannot write a message that sounds like it noticed something specific about one person. The honest trade-off is this: tools save you time on steps 2 through 5 of a campaign, mail merge, scheduling, follow-up timing, but step 1, the actual observation that makes a message land, still has to come from a human reading the person's profile and recent activity. If you are comparing options, I broke down what each platform does well in the best LinkedIn automation tools compared, including which ones are built for volume versus which ones are built for quality control on smaller lists.

If your outreach lives across both LinkedIn and email, the same five-piece structure carries over, and you can shortcut the writing with ChatGPT prompts for cold email outreach built to get replies, then adapt the output down to LinkedIn's character limits rather than starting from a blank page both times. There are also general-purpose ChatGPT prompt templates for everyday work tasks if you want to build your own message variations faster without reinventing the structure every time.

What happens after they say yes

A reply is not a sale, it is permission to send the next thing, and that next thing should be as low-friction as the message was. If your plan is to push a reply straight into a 45-minute discovery call, you will lose a chunk of the people who would have taken a smaller step first. Sending a short resource, a one-page breakdown, or even a direct link works better for warming someone up than asking for calendar time immediately. If that resource lives on a dedicated page rather than your homepage, response and conversion both tend to improve, which is covered in when to use a landing page in your marketing funnel.

Frequently asked questions

What is the ideal length for a LinkedIn cold message?

Under 500 characters for a first message and under 300 characters for a connection request note, since that is LinkedIn's hard limit on notes. Shorter almost always outperforms longer because it respects the reader's time and gets fully read on mobile without a "see more" tap.

Should I ask for a call in the first message?

No. Asking for 15 or 30 minutes in your opening message is the single most common reason cold messages get ignored, because it asks for a large commitment from someone who has not decided they trust you yet. Ask for something small first, like permission to send a resource or a one-line answer to a question.

Does personalisation increase reply rates?

Specific personalisation does, generic merge-field personalisation often does not, and can read worse than no personalisation at all. Mentioning a real, recent, checkable detail about the person or their company works far better than inserting their job title or company name into a template sentence.

How many follow-ups should I send after a cold message?

Two is usually enough: one around 10 to 14 days after the first message, and one final message a few weeks after that which explicitly offers to check back later instead of pushing again. More than two follow-ups on LinkedIn tends to read as pressure rather than persistence.

Published and maintained by the Lilach Bullock team, covering marketing, AI and business growth.
Your buyers are asking AI who to use. Does it say you?

See for free whether ChatGPT, Claude, Perplexity, Gemini and Google name you, and get the plan to become the answer.

Check my AI visibility →
Sundays only

Get the Sunday newsletter.

One email a week. AI experiments, marketing tactics, and the workflows Lilach is building right now in her own business.

Subscribe free

Let’s get your marketing running on AI.

Book a free 30-minute call

We figure out what you need, where AI fits in, and what working together would look like.

Book the call →

Or take the 30-second calculator

You’ll see the hours and the money quietly leaking out of your week, and the three workflows worth building first.

Take the calculator →

Or grab the free AI resource library

Prompt packs, templates, checklists, and swipe files. The exact tools I build for paying clients. Yours, free.

Get the library →
Keep reading

More from the blog.