The question comes up in almost every conversation I have with a coach or a creator: "If I automate my DMs, am I putting my account at risk?"
It's a good question, and most of the answers out there are bad. Some say "zero risk, it's official." Others say "never automate anything." Both are wrong, because they're talking about the tool, while Instagram is looking at behavior.
Here's what actually triggers a restriction, what changes nothing, and why following up on your old conversations is very different from cold-messaging strangers.
What Instagram actually watches
Instagram doesn't have an automation detector. It watches signals, and those signals describe a way of behaving.
Volume over a short window. Fifty messages sent in ten minutes don't look like anything a human does. The platform doesn't need to know which software produced them to spot the abnormal pace.
Identical repetition. The same text sent to a hundred people is the simplest signal to catch, and the oldest. It's also the one cheap tools produce by design. Meta's spam policy names it plainly.
Writing to people who never talked to you. This is the most important point by far. Opening a conversation with a stranger is the move the platform watches most closely, because it's exactly what a spammer does.
Reports. Someone who gets an unwanted message can report it. A handful of reports is enough to push an account into a monitored category, whatever tool you use.
Notice that nothing on this list is about using software. A human who sent a hundred identical messages to strangers in an hour would be sanctioned exactly the same way.
The difference nobody explains
There's a very clear line between two actions that everyday language lumps together as "automation."
Replying to someone who wrote to you. The person opened the conversation, they know you, they may even be waiting for a next step. In the platform's eyes, you're maintaining an existing relationship.
Soliciting someone who never wrote to you. You open a conversation that didn't exist. That's where most of the risk sits.
This line isn't a fine distinction. It separates a calm account from an account on probation. And it explains why following up on your own history is a much safer operation than a prospecting campaign.
Your old leads wrote to you. Sometimes six months ago, sometimes two years ago. The conversation already exists in your inbox. Going back to it doesn't create anything new. It picks up a thread the person opened themselves. We go deeper on why that history is where the money is in following up on old conversations.
The three mistakes that cost an account
On the accounts I've seen get restricted, the causes almost always come down to three moves.
Buying a list and messaging everyone. It's the fastest road to a restriction. These people don't know you, many of them will report you, and the volume shows up immediately.
Sending the same message at scale. Even when you're writing to people who know you, a hundred carbon copies of the same text form a pattern. A message that reads like what you actually write, with the words you actually use, doesn't.
Sending everything at once. A queue that empties in a few minutes is the crudest signal there is. Messages spread through the day simply look like someone working through their inbox. We break down the pacing side in detail in how often to follow up on Instagram.
There's a fourth mistake, quieter and so more vicious: viewing profiles you don't know in bulk. Many tools do it to enrich their lead data without telling you. That repeated reading is a signal too, and it happens before the first message is even sent.
What makes a follow-up safe
A follow-up that doesn't put your account at risk meets four conditions, and each fits in one sentence.
It goes to someone who already wrote to you. That's the main condition, the one that makes the difference between a reply and a solicitation.
It's written in your voice. Not a copied template or a sales line, but the message you'd have written yourself if you'd had time to scroll back through two years of conversations. Something like:
Hey Marcus, you mentioned in March you wanted to wait until your launch was done. How did it go?
It goes out at a human pace, spread across the day rather than in a burst.
And above all, a human approves before it's sent. That's the safeguard that catches everything else: if the message is off, out of context, or addressed to the wrong person, someone notices before it goes out. We wrote a full piece on why a human in the loop matters.
That's the principle Kord runs on. Kord is Ficiel's AI assistant: it rereads your conversations, finds the leads who went quiet, drafts follow-ups in your own voice, and a human always approves before anything is sent.
Personal account or professional account
A practical point, because it causes a lot of confusion and has nothing to do with bans.
On a personal Instagram account, the conversation history available to outside tools is very limited. It's not a bug or a sign of an empty inbox. It's a difference in nature between the two account types. A professional account gives access to a much more complete history.
If you sell through DMs and your account is still personal, switching to professional is free and takes a couple of minutes. Do it before anything else, or you'll wrongly conclude there's nothing to recover in your inbox.
One more thing to know: indexing a large history doesn't happen all at once. The first conversations show up quickly, and the rest fills in gradually over a few hours. Seeing only a handful of conversations in the first minutes means nothing.
If your account is already restricted
A restriction is almost never permanent, and it rarely looks the way people imagine.
Most of the time, you get no notification at all. Your messages seem to go out normally, but they no longer arrive, or only reach some recipients. It's the most common form, and the most confusing, because nothing tells you there's a problem.
Sometimes sending is blocked with an error message, often for a few hours to a few days. That's the most visible form and, oddly, the least serious: it lifts on its own.
In both cases, the fix is the same and it's simple: stop all automated sending for several days. Keep using the account normally from your phone, reply by hand, post as usual. What the platform sees next is a return to ordinary behavior.
What you absolutely shouldn't do is switch tools and start again right away at the same pace. The account is already being watched. The same behavior will produce the same result, faster.
And when you start again, start with your own conversations. Replying to people who wrote to you is precisely the opposite of the signal that caused the restriction.
So, should you be afraid?
No. But you should know what you're doing.
Fear of getting banned is fed by two opposite camps: people selling mass prospecting tools, who downplay the risk, and people selling done-for-you manual services, who dramatize it. The reality is more boring than both pitches. It's also written in Meta's own rules: its messaging policy treats replying to people who contacted you as the normal behavior of a business account.
Don't confuse this with ordinary acquisition tools, the ones triggered by something the lead does: a comment, a reply to a story. Those tools work on the right side of the line, because the person spoke first.
If you write to people who wrote to you, in your own words, at a normal pace, and a human reads before anything is sent, you're doing the most ordinary thing there is. It's what any business owner does when they answer their customers.
If you buy a list and send a thousand identical messages to strangers, no tool in the world will protect you, and the problem won't be technical.
FAQ
What is Ficiel?
Ficiel is an AI inbox for coaches and setters: it surfaces old Instagram, WhatsApp and LinkedIn conversations worth following up, drafts the message, and nothing is sent without your approval. You connect your own account, just like on your phone, and you approve each message before it goes out.
Does using an automation tool get an Instagram account banned?
Not by itself. What triggers a restriction is behavior: abnormal volume over a short window, identical messages in series, and above all messages to people who never wrote to you. An account that replies to its own conversations at a human pace doesn't send those signals.
Is it risky to follow up with someone who messaged me a year ago?
It's the safest use there is. The person opened the conversation themselves: you're picking up an existing thread, not creating a new one. That's very different from writing to a stranger.
Is there a safe daily number of DMs?
No public threshold exists, and be wary of any article that gives you a precise one: these limits aren't published and vary with the age and history of the account. The right move isn't to aim for a number. It's to spread your messages out and keep a volume comparable to what a person would handle by hand.
What happens if a message goes to the wrong person?
With Kord, the question plays out differently: nothing goes out until a human has approved it. That's exactly what this step is for, and it's what separates a prepared follow-up from an automatic send.
Do I need a professional account?
To recover a complete conversation history, yes. On a personal account, the number of accessible conversations is very small, which gives the misleading impression there's nothing to follow up on. Switching to professional is free.
