How stop-on-reply works across channels

What exactly stops?

Every step of that prospect's sequence that has not happened yet. Concretely:

• Scheduled follow-ups are cancelled — the second email, the third email, and anything after them.
• Pending actions on the other channels are cancelled — the LinkedIn connection request or message, the Instagram DM, the WhatsApp message.
• The prospect's enrollment ends in that campaign and is marked as replied.

What does not stop: the campaign itself keeps running for everybody else, and the same person can still be enrolled in a different campaign later. Stop-on-reply is a per-person decision, never a per-campaign one.

Stop-on-reply is a campaign setting and it is on by default. You can turn it off for a campaign where continuing after an answer is genuinely what you want, but for cold outreach leaving it on is what keeps the sequence from talking over the person you just started a conversation with.

What counts as a real reply?

A message a person actually wrote back to you, on any of the channels the sequence uses.

A message someone sends you out of the blue — one that is not an answer to anything in the sequence — also arrives in your inbox, but it is the reply on the prospect's own thread that ends their sequence.

Why doesn't an out-of-office stop my sequence?

Because an out-of-office is a mail system answering on someone's behalf. It tells you the person is away, not that they said no — and stopping there would quietly kill a lead who never read anything you sent.

Automatic replies are recognised from three kinds of evidence: the hidden markers mail systems attach to auto-responders, the shape of the subject line, and the wording of the message. The markers are language-independent, which matters because phrase matching alone can only ever cover the languages it was written for.

The important part: an automatic reply is filed as an automatic reply, not deleted. The message stays in your inbox and stays in the prospect's history — it is simply recorded as the automatic thing it is, rather than as a reply from the person. Nothing is thrown away to make the sequence keep running, so you can always see exactly what arrived and when. That is a deliberate design choice: a record you can audit is worth more than a tidy one.

The same distinction is why an out-of-office never inflates your reply rate. It is excluded from reply counts and from any step in your sequence that waits on a reply, so a week of holiday autoresponders cannot make a campaign look like it is working when it is not.

See also: does an out-of-office reply stop my campaign?

What if a reply is matched to the wrong step?

It shouldn't be, and this is worth explaining because it is the sort of thing that quietly ruins your reporting.

A sequence keeps a pointer to whatever it is going to do next, and that pointer moves forward the instant a step is sent. A reply, meanwhile, arrives some time later — minutes, hours, sometimes days. If the reply were labelled with wherever the pointer had got to, it would be credited to a step the prospect had never received. The opener that actually earned the answer would read as a zero, and a follow-up that was never sent would read as your best-performing step.

So the step a reply refers to is resolved from what was actually sent — the sending history for that prospect, and the thread the reply arrived on — never from where the sequence pointer happens to be. The same rule applies to bounces. In practice this means your per-step reply and bounce numbers describe the messages that really produced them, which is the only version of those numbers worth acting on.

Can I resume a sequence that stopped?

Yes, and nothing is lost by doing it. There are two ways, and both are non-destructive:

From the campaign

Open the campaign, go to the Prospects tab, find the person, open the actions menu on their row and choose Resume sequence.

From the inbox

Open the message and relabel the thread as Auto-Reply / Out of Office. The sequence picks up again on its own, and the message stops counting toward your reply rate.

Two guarantees either way:

• Nothing is re-sent. The prospect rejoins at the next step they have not already received. An email that has gone out once can never go out twice, however many times a person is stopped and resumed.
• Nothing goes out immediately. The next step is scheduled at the campaign's normal pace, inside its normal sending window — resuming never produces a burst.

Resuming also keeps the record intact: the message that caused the stop is re-filed as an automatic reply rather than removed, so your history still shows what arrived.

Why did a sequence keep sending after a reply?

Almost always one of these, in rough order of how often it turns out to be the cause:

• The reply arrived after the next step had already gone out. Stop-on-reply cancels steps that have not happened yet; it cannot recall a message that was already sent. If a follow-up went out an hour before the answer landed, that follow-up was already gone.
• What arrived was an automatic reply. By design, that leaves the sequence running. Check the thread's label in the inbox.
• The person replied from a different address, or from a personal account rather than the one you wrote to, so the answer is not on the prospect's thread.
• Stop-on-reply is turned off for that campaign.
• The person is enrolled in more than one campaign. A reply ends the enrollment it belongs to; if they are also in a second campaign, that one has its own steps. Suppression lists are the tool for taking someone out of everything at once.

Does a reply on one channel stop the other channels?

Yes. This is the whole point.

The channels that stop together are cold email, LinkedIn, Instagram and WhatsApp, plus any multichannel sequence that spans several of them. A reply on any one of them stops the pending steps on all the others for that prospect. There is a single shared list of which channels stop-on-reply covers, so when a new channel is added it is covered everywhere at once rather than being remembered in some places and forgotten in others.

Worked example. A prospect is in a multichannel sequence: email on day 1, LinkedIn connection request on day 3, follow-up email on day 5, Instagram DM on day 8. They answer the day-1 email on day 2. The LinkedIn request, the follow-up email and the Instagram DM are all cancelled, and the conversation continues in your Unified Inbox on the channel they chose.

More on running channels together: multichannel outreach.

Does an Instagram reaction count as a reply?

No. A reaction — a like, or an emoji added to your message — is not a message, so it is not a reply and the prospect stays in the sequence.

This is not a gap we are planning to close. A reaction carries no content and no intent you can act on; treating one as an answer would stop a sequence for someone who has told you nothing, and you would never hear from them again. If the person wants to talk, they will write something, and that stops the sequence immediately.

Worth knowing alongside this: reacting to a direct message is not something Instagram itself supports, so a reaction on that channel is not a thing you will send or receive there in the first place. On the channels where reactions do exist, the rule above is what applies. For the other Instagram specifics, see Instagram sending limits and managing your Instagram inbox.

What happens if the prospect asks to be removed in the reply?

They are opted out, not merely stopped — and the two are different things.

A reply asking to be removed is recognised as an opt-out before it is filed as an ordinary reply. The prospect is unsubscribed, added to your suppression list so future campaigns do not reach them, and their sequence ends. A written request is exactly as strong as clicking a link; you do not have to spot it yourself and act on it manually.

You choose how the opt-out is offered in your campaign emails — a link, or a reply instruction pointing at your own sending address — and you write the sentence yourself so it sounds like you. What you cannot do is remove the opt-out entirely: the outcome is guaranteed, only its shape and wording are yours. See the opt-out link in campaign emails.

Can an AI agent handle this for me?

Partly, and the line is worth being exact about. A connected agent — Claude, ChatGPT, Cursor, Codex, OpenClaw, Hermes Agent, or any agent that speaks MCP — can build, launch, pause and resume campaigns on every channel, create and enroll prospects, verify email addresses, add people to a suppression list, configure warmup and read your stats, all in plain language. There is no tool that lets an agent read your inbox threads or send a reply, so answering a real person stays yours.

And the framing that never comes off: the agent never sends a message, DM or invite itself, and can never raise a limit. Launching a campaign only writes it and hands it to WarmySender's scheduler, which paces every cold email, LinkedIn, Instagram and WhatsApp action inside safe caps and the gradual ramp regardless of whether a person or an agent triggered it. Connecting and disconnecting accounts stays in the app.