Why did my prospect get the same LinkedIn message twice?
Because a sequence step could deliver its message and then fail to record that it had done so — and a step with no record of having run looks, to the next pass, like a step that still needs to run. That is what produced the repeats. It was not something you configured wrong, it was not a duplicate prospect in your list, and it was not your LinkedIn account. This has been fixed: every message is now recorded the moment it is delivered, before anything else is updated, so a follow-up step can no longer send a message it has already sent.
If you are reading this because a prospect just received the same message several times, here is the short version:
- It has stopped. A step can only send once per person now, and that is enforced at the point of sending rather than checked afterwards.
- Nothing about your campaign needs rebuilding. Your sequence, your audience and your connected account are all fine, and there is no setting for you to change.
- Your LinkedIn account is not at risk from it. The repeats all went out inside your account's normal daily pace — they used up your sending allowance, but they never pushed past it.
- There is a good way to handle the person who received them. See what to say to the prospect below.
WarmySender is an agentic outreach platform covering Cold Email, Email Warmup, LinkedIn, Instagram and Multichannel sequences, plus real-time email verification. This guide covers a LinkedIn Outreach question: repeated messages on a sequence step, and the protections that stop them.
What does a duplicate actually look like?
The same message, word for word, appearing several times in one LinkedIn conversation — usually spaced hours apart, and usually from a single step in the middle of your sequence. It is unmistakable when you see it, but it is easy to miss if you are not looking, because it lives inside a conversation you are not opening every day.
The tell-tale signs, in order of how obvious they are:
- An identical message repeated in the thread. Not a variation, not the next step of your sequence — the same text, the same placeholders filled in the same way.
- Regular gaps between the copies. They tend to arrive at intervals rather than in a burst, because each one went out through your account's normal safe pace.
- It is one step, not the whole sequence. The other steps in your sequence behave normally; it is a single step that keeps coming back around.
- It often happens to someone who already replied. That is the version of this people notice first, because a reply sitting between two identical follow-ups is impossible to miss.
Where to look: open the campaign, go to the Prospects tab, and open the conversation for the person in question. The conversation view shows every message sent to that person from that account in order, so repeats are visible at a glance.
Why did it happen?
Delivering the message and writing down that it was delivered were two separate actions, and the second one could fail on its own. When it did, the message had genuinely reached the prospect — but nothing anywhere said so. The next time the sequence looked at that prospect, it found a step that appeared not to have run yet, and did the obvious thing: it ran it.
Then the same thing happened again. And again. Each pass delivered a real message and failed to record it, so nothing accumulated that could break the loop. That is why the counts got as high as they did in some conversations rather than stopping at two.
Two things made it land hardest on the people you would least want it to:
- Prospects who had already replied were the most exposed. The bookkeeping that failed was most likely to fail on someone whose conversation had already moved on — which is exactly the person for whom a repeated cold follow-up reads worst.
- The step looked like it had failed, so it was retried. From the outside, a step that never recorded a success is indistinguishable from a step that never ran. Retrying is normally the right instinct — a genuinely failed step should be tried again — and here it was the wrong one.
The important part, and the reason this is worth writing down: the check that was supposed to prevent a repeat depended on the very record that failed to save. A safeguard that reads a note nobody managed to write is not a safeguard. That is the thing that changed.
What has changed so it cannot happen again?
The delivery is now recorded first, the moment the message reaches the prospect, and everything else happens afterwards. Sending a LinkedIn message cannot be undone, so it is now treated as the thing that must be written down before anything else is touched. Once that record exists, no later step can remove it, and no retry can talk its way past it.
In practice that means four protections, each of which would be enough on its own:
- Record the delivery, then update everything else. The message going out is the first thing written down. If any later bookkeeping runs into trouble, the record of the delivery is already safely in place and the step counts as done.
- One send per step, per person — enforced, not just checked. A second attempt at the same step for the same person cannot be recorded at all, so even a step that is retried for some unrelated reason cannot produce a second message.
- A retry that finds a delivery already recorded moves on instead of repeating. Where the old behaviour was "no record, so run it", the new behaviour is "there is a record, so continue to the next step".
- A step that keeps failing stops rather than looping forever. If a step genuinely cannot complete after several attempts, it stops for that person and is flagged, instead of retrying indefinitely. You will see this on the campaign's Prospects tab with a plain-English reason.
None of this changes your sequence, your pacing, or your account's daily limits. The safe sending pace for a LinkedIn account is unchanged — the fix removes a repeat, it does not add throughput.
How does "Stop on Reply" fit into this?
"Stop on Reply" is a separate protection — it stops the rest of a sequence once someone answers — and it was never the cause of duplicates. It is worth understanding because it is the single most useful safety net you have on a campaign, and because leaving it off is a legitimate choice that some campaigns are deliberately built around.
What it does when it is on (the default): as soon as a prospect replies, the remaining steps of your sequence stop for that person. No further follow-ups, no further connection nudges. The conversation is now a conversation, and the sequence gets out of the way.
What it does when it is off: your follow-ups keep going out on schedule even after someone answers. That is genuinely what some people want — a two-part introduction where the second message adds context regardless of what came back, or a nurture sequence where a short "thanks, I'll take a look" should not cancel the rest. It is not a misconfiguration, and nothing on your campaign is broken because of it.
| Stop on Reply ON (default) | Stop on Reply OFF | |
|---|---|---|
| Someone replies mid-sequence | Remaining steps stop for that person | Remaining steps keep sending on schedule |
| Best for | Most outreach — a reply is the goal, so the sequence has done its job | Sequences designed to be read in full, or where a short acknowledgement is not the answer you are after |
| Risk to watch | None in normal use | Someone who says "not right now" still receives the rest, which can read as tone-deaf |
| Related to duplicate messages? | No | No — a repeated message is one step sending twice, which is a different thing entirely |
Where do I find and change it?
Open the campaign and look at the Overview tab. If Stop on Reply is switched off on a running campaign, a short note says so there, so you never have to go looking to find out. The full list of safety rules — Stop on Reply, Stop on bounce, Stop on unsubscribe — is on the campaign's Settings tab, and you change any of them with Edit Settings on that same tab, which opens the campaign editor. The setting applies to that campaign only; your other campaigns keep their own.
If you are not sure which you want, leave it on. A reply is almost always the outcome you were sending for, and stopping there is the polite default.
What do I say to the prospect who got the repeats?
Say something short, human and unbothered — then move the conversation forward. The single worst response is silence, because it leaves the other person deciding for themselves whether you are careless or automated. A one-line acknowledgement resets it, and in practice most people find it disarming.
Concrete examples you can adapt, depending on what happened:
- They received several copies and have not replied: "Apologies for the repeats — that was on my end, not you. Ignoring all but the first one is entirely reasonable. If the original is of any interest, I'm happy to pick it up from there."
- They replied and then got more copies: "Sorry about the extra messages after your reply — completely my side. Coming back to what you actually said…" and then answer them properly.
- They pointed it out, with humour or without: "Fair. It went out more times than anyone needed. Fixed now — and thanks for being good-natured about it."
- They asked to be left alone: stop, and say so in one line. "Understood — you won't hear from me again. Apologies for the noise." Then remove them from the campaign so nothing further goes out.
Keep it to one or two sentences. A long apology draws more attention to it than the duplicates did, and turns a small annoyance into the subject of the conversation.
What should I check on my side?
Nothing is required — but three quick checks will tell you the scope of what happened and leave your campaign tidy. All three take a minute and none of them change anything about how your campaign runs.
- Look at the affected conversations. Campaign → Prospects tab → open the person. You will see every message sent to them in order, so you can tell exactly how many copies went out and when the last one was.
- Check whether anyone who replied is still in the sequence. A prospect who has answered should be showing as replied. If someone answered and is still moving through steps, and you want that to stop, switching Stop on Reply on for that campaign is the setting that does it.
- Remove anyone who asked to be left alone. On the Prospects tab you can stop or remove an individual person, which takes them out of the sequence immediately without affecting anyone else.
What you do not need to do: rebuild the campaign, re-import your audience, disconnect and reconnect your LinkedIn account, or duplicate the campaign to get a clean copy. None of those change anything here, and reconnecting an account has its own short settling period, so it costs you time for no benefit.
How do I tell a duplicate from something that only looks like one?
Three ordinary situations get mistaken for a duplicate, and each has a different tell. It is worth a quick look before assuming a repeat, because the fixes are completely different.
- The same person imported twice. If a prospect appears more than once in your audience, each copy runs the sequence independently, so they receive everything twice, in step. The tell: the whole sequence repeats, not one step, and the person appears twice in your Prospects list. The fix is to remove the duplicate entry.
- The same person in two campaigns. Two campaigns running against overlapping lists will both message the same person. The tell: the messages are different — they come from different sequences. Exclusion lists are what prevent this; see setting up campaign sequences.
- A follow-up that reads similarly to the previous step. If two steps in your sequence open the same way, they can look like the same message at a glance. The tell: read past the first line — the body differs. Nothing to fix, though it is often worth varying the opener.
A genuine repeat is a single step, identical text, to one person, more than once. Anything else on this list is a different problem with a different answer.
What this did not affect
It was confined to repeated messages on a LinkedIn sequence step. Specifically:
- Your cold email campaigns are unaffected. Email sending, follow-up threading and your sending schedule are separate and were never involved.
- Your email warmup is unaffected. Warmup runs on its own and does not touch campaign sending.
- Your LinkedIn account's safe pace was never exceeded. Every repeat went out through the same daily and hourly limits as any other action, so a duplicate consumed part of your allowance but never pushed your account past it.
- Connection requests were not repeated. LinkedIn itself will not accept a second invitation to someone who already has one pending or connected, so the invite steps were never able to duplicate.
- Nothing was lost. No prospect was dropped, no campaign was reset, and no reply was missed.
Frequently asked questions
Why did my prospect receive the same LinkedIn message twice?
Because a sequence step could deliver its message and then fail to record that it had done so. With no record of the message having gone out, the next pass at that prospect saw a step that appeared not to have run yet and ran it, which sent the same message again. It was not caused by anything you configured, by a duplicate in your list, or by your LinkedIn account. It has been fixed: every message is now recorded the moment it is delivered, before anything else is updated, so a step can no longer send a message it has already sent.
Is it still happening?
No. Each message is recorded at the point of delivery, and a second send of the same step to the same person can no longer be recorded at all, so it cannot be produced. On top of that, a retry that finds a delivery already recorded now moves on to the next step instead of repeating, and a step that genuinely cannot complete after several attempts stops for that person and is flagged rather than looping. Nothing is required from you for any of that.
Did I do something wrong in my campaign settings?
No. There was no setting that caused this and no setting that would have prevented it. In particular it was not caused by Stop on Reply being switched off — a repeated message is one step sending twice, which is a different thing from a sequence continuing after someone answers. Your sequence, your audience and your connected account do not need changing, and the campaign does not need rebuilding.
Why did it hit people who had already replied?
Because the bookkeeping that failed was most likely to fail on a prospect whose conversation had already moved on — which is exactly the person for whom a repeated cold follow-up reads worst. That is why the version people noticed first was a reply sitting between two identical follow-ups. Both halves of that are now closed: the delivery is recorded before anything else is touched, and a prospect who has replied is recognised as replied everywhere in the sequence, not just in the place that happened to be checked.
Is my LinkedIn account at risk because of the extra messages?
No. Every repeat went out through the same daily and hourly limits as any other action on your account, so the duplicates used part of your normal sending allowance but never pushed your account past its safe pace. Account safety limits are never relaxed to make anything run faster, including recovery from a problem like this one.
What does "Stop on Reply" actually do, and should I have it on?
When it is on — which is the default — the rest of your sequence stops for a person as soon as they reply, so they get no further follow-ups. When it is off, your follow-ups keep sending on schedule even after someone answers. Off is a legitimate choice for sequences meant to be read in full, or where a brief acknowledgement is not the reply you were after. If you are unsure, leave it on: a reply is usually the outcome you were sending for. It is not related to duplicate messages either way.
Where do I change Stop on Reply?
Open the campaign, go to the Settings tab, and use Edit Settings to open the campaign editor, where Stop on Reply sits with the other safety rules. If it is switched off on a running campaign, the campaign's Overview tab now shows a short note saying so, so you can see the state without going looking for it. The setting applies to that campaign only — your other campaigns keep their own.
What should I say to the prospect who received the repeats?
Something short and human, then move on: "Apologies for the repeats — that was on my end, not you. If the original is of any interest, happy to pick it up from there." If they had already replied, acknowledge it in one line and then answer what they actually said. If they asked to be left alone, say "Understood — you won't hear from me again", and remove them from the campaign. Keep it to a sentence or two; a long apology makes the duplicates the subject of the conversation.
How do I stop messages to one specific person right now?
Open the campaign, go to the Prospects tab, find the person and stop or remove them there. That takes them out of the sequence immediately and does not affect anyone else in the campaign. If you want replies to stop the sequence for everyone from now on, switch Stop on Reply on for that campaign instead.
Did this affect my cold email campaigns or my warmup?
No. It was confined to repeated messages on a LinkedIn sequence step. Email sending, follow-up threading, sending schedules and email warmup are separate and were never involved. Connection requests could not duplicate either, because LinkedIn will not accept a second invitation to someone who already has one pending or is already connected.
How do I know it was a repeat and not the same person imported twice?
A genuine repeat is one step, identical text, to one person, more than once — the rest of the sequence behaves normally. If the same person was imported twice, the whole sequence repeats rather than one step, and they appear twice in your Prospects list; removing the duplicate entry fixes it. If two campaigns are running against overlapping lists, the messages will be different because they come from different sequences, and an exclusion list is what prevents that.
Related reading
- Why some prospects stopped getting follow-ups after accepting — the opposite symptom: a message that has not gone out yet, and why
- How to set up campaign sequences: delays and conditions — steps, waits and branching, including where safety rules apply
- LinkedIn sending limits and account tiers — the daily and weekly pace your account sends at
- Why your new LinkedIn account starts slow — the ramp a newly connected account follows
- Reconnecting a LinkedIn account and your campaigns — what a reconnect does and does not reset
- LinkedIn safety FAQ — invitations, messages, limits and keeping an account healthy
- Full documentation — Cold Email, Email Warmup, LinkedIn, Instagram and Multichannel guides
Still have questions? Email [email protected] — we respond within 24 hours on business days.