A LinkedIn outreach strategy for B2B that the platform allows

Ignored connection requests count against your account. That single documented fact rules out most of the advice written about this channel.

The advice on LinkedIn outreach splits into two camps. One says volume, automate, accept a low reply rate. The other says relationships, patience, no pitching. Both are describing a strategy for a platform that has its own opinion, and the platform's opinion is enforced.

Worth starting there, because it changes the arithmetic that everything else rests on.

The constraint that decides your strategy

LinkedIn restricts accounts that send too many invitations, and it publishes the reasons. Its help documentation lists three: you have 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, or an excessive number suggests the use of an automation tool.

Read the second one again. Ignored invitations count against you. That is not a spam rule, it is a relevance rule, and it means the cost of a badly targeted request is not zero. You are not spending only your time, you are spending your account's standing.

The company is also deliberately unhelpful about the exact numbers. The same documentation states that support cannot disclose the type or reason for a restriction, and that restrictions cannot be removed on request. Most clear within a week on their own, and a withdrawn invitation cannot be re-sent to the same person for up to three weeks.

So any strategy built on a specific daily limit you read somewhere is built on a number the platform does not publish and can change. The stable strategy optimises acceptance rate instead, because acceptance rate is the thing the restrictions are actually measuring.

What that means in practice

If ignored requests are a cost, then sending fewer, better-aimed requests is not a softer approach. It is the approach with the higher ceiling, because it protects the ability to keep sending.

• Send to people who will recognise why. A shared client, a shared event, a specific post of theirs. Recognition is what converts a request into an acceptance.
• Stop sending to whole titles. "All operations directors in the Gulf" is a search result, not a target list. It produces exactly the ignored-invitation pattern the policy names.
• Withdraw pending requests on a schedule. Old pending invitations sit in the ignored column. Clearing them monthly keeps the ratio honest, remembering the three-week wait before that person can be invited again.
• Treat automation as a risk, not a shortcut. The documentation names suspected tooling as its own trigger, separate from volume. That is a policy risk you carry, not a tactic.

The sequence that works, and why it is boring

Four steps, and the important thing is that the first two contain no pitch at all.

1. Be visible before you arrive. Comment on something they wrote, with something specific. Cheap, and it makes the request recognisable rather than cold.
2. The request itself, with a reason. One sentence naming why them specifically. No offer, no meeting request.
3. Wait. A pitch arriving within a minute of acceptance is the single most reliable way to teach someone that you are running a sequence.
4. Ask a question you could not answer yourself. Not "would you like a call". Something about their situation that shows you looked, and that has a short answer.

Step four is where most sequences fail, because the message is written to serve the sender's calendar rather than the reader's interest. If the message would work equally well sent to two hundred people, it will perform like a message sent to two hundred people.

The reply rate is not a copywriting problem. It is a targeting problem wearing a copywriting costume.

How many touches, honestly

We do not publish a magic number, because ours comes from our own accounts and would be a fabricated benchmark if we presented it as an industry figure. What we will say is the shape: replies cluster early, the tail is long and thin, and the second message does more work than any message after it.

The practical rule we use is three touches on LinkedIn, then stop, and record why. A prospect who has ignored three well-aimed messages is data, not a failure, and pushing to five is how the ignored-invitation ratio quietly deteriorates.

Where LinkedIn is the wrong channel

Two cases, stated plainly because pretending otherwise wastes months.

The first is any buyer who does not use the platform. Plenty of operations, manufacturing and family-business decision makers across our markets have a dormant profile and a very active phone. Outreach there is a phone and an introduction, and no sequence changes that.

There is a third case worth naming: when the honest answer is that you should be doing this in-house rather than buying it, which we set out in outsourced BD versus hiring. The second is when the constraint is not reach at all. Khaled Badr's write-up on using direct LinkedIn outreach as a growth channel without ads is a good companion here, because it comes at the same channel from the owner's side rather than the sales team's, and it is blunt about what it costs in time.

The measurement that keeps you honest

Track three things and you will know within a month whether this is working:

• Acceptance rate. The health metric, because it is what the platform's restrictions respond to. Falling acceptance is an early warning, not a vanity number.
• Reply rate on accepted connections. Measures the message, separated from the targeting.
• Meetings that met your qualification bar. Not meetings booked. The distinction is the whole difference between activity and pipeline.

Those three sit at the top of what we report on, and they are deliberately narrower than the dashboards most outreach tools produce. What we commit to is set out in the deliverables, and the awkward questions people ask before starting are answered in the FAQ. If acceptance is healthy and replies are not, rewrite the message. If acceptance is poor, no message will save it, and the list is the thing to fix. Most teams do this the wrong way round and spend six weeks on copy.

The honest summary

LinkedIn outreach works, slowly, when it is aimed. It stops working, quickly, when it is scaled past the point where the people receiving it recognise why they were chosen. The platform enforces that boundary itself, which is unusual and, if you are being fair about it, reasonable.

If the process behind the outreach is undefined, none of this survives contact with a busy month. That is the argument we make in system problem, not effort problem, and it applies to this channel more than most, because outreach is the easiest activity in the company to do busily and pointlessly.