The question always comes in the same shape: how many messages can I send per day without risk? It's a fair question. It's the first objection of anyone who sells through DMs. And it doesn't have the clean number everyone is looking for.
We've already covered what actually triggers an account restriction. This article picks up from another angle, more practical and rarely discussed: at what pace, and with what mechanics, should a tool send your follow-ups so they look like a person working through their inbox rather than a bot emptying a queue?
The number you want doesn't exist, and the reasons it doesn't are useful. What's left is what the platform looks at instead, the settings that actually protect an account, and two questions that are enough to size up any tool on this front.
Why nobody can give you a number
Instagram doesn't publish a sending limit for the DMs of a regular account. That silence isn't an oversight. A published limit becomes a target, and a known threshold gets gamed. The Community Guidelines describe behaviors, not quotas, which leaves the platform free to judge case by case.
The contrast with WhatsApp tells you a lot. For the business infrastructure that high-volume sending tools plug into, Meta does publish messaging tiers, with numbers and the conditions to move up. When a limit exists and the platform wants it respected, it writes it down. For the Instagram DMs of a regular account, it hasn't.
So be wary of any content that gives you a precise figure. These limits vary with the age of the account, its history, and signals nobody outside Meta can see. A number presented as universal is, at best, an average someone observed on their own accounts.
What the platform watches is a rhythm, not a total
If the total isn't the right indicator, the pattern is. Fifty messages spread across a workday by someone who is also replying to their conversations look nothing like fifty messages fired off in eight minutes, even though the total is the same. What separates the two is human irregularity. A person takes more or less time depending on the message, gets interrupted, comes back.
Three signals matter more than volume.
The first is spacing. A perfectly regular cadence is a signature on its own: one message every sixty seconds exactly, for an hour, has no chance of passing for a human.
The second is repeated content. Meta explicitly lists the same text sent in series as spam.
The third, and by far the heaviest, is who you're writing to. Writing to someone who already wrote to you is a reply. Meta even describes that gesture in its messaging policy as the expected behavior of a business account.
There's a fourth factor that weighs a lot and rarely gets mentioned: how recently the account was connected. An account that was just plugged into a tool is at its most fragile, because the platform sees a new session from an unfamiliar device. Sending a lot in those first days is the most expensive thing you can do, and it's exactly when impatience runs highest.
The four settings that protect an account
They're short, and they don't depend on any specific vendor.
Variable spacing between two messages. The interval should change from one send to the next. A metronome doesn't look like anyone working through their inbox.
An hourly cap per account, on top of a daily cap. A daily quota alone lets you send everything in the same half hour, which recreates the exact pattern you were trying to avoid.
Messages that differ from each other. A template with the first name swapped is still a template. A message that picks up what the person told you looks like no other message, and it's the only reliable way to stay clear of repeated-content detection. Compare:
Hey Sarah, just checking in to see if you're still interested in coaching!
Hey Sarah, last time you said you wanted to get back to it after your move. Did the new place settle down?
The first one could go to anyone. The second could only go to her.
Go easy on recently connected accounts. The first days call for fewer requests, especially anything that isn't messaging. Reading profiles or posts in bulk looks far more like a scraper than writing to someone does.
None of these four settings is about automating or not. A human who breaks them gets sanctioned the same way, and we've detailed elsewhere what actually triggers an account restriction.
What Ficiel actually does
The follow-ups you approve from the dashboard go out one at a time, spaced a few minutes apart, under an hourly cap per account. The spacing is drawn at random within a range rather than fixed, for the reason above.
When the cap is reached, the follow-up is refused on the spot, with the reason and how long to wait before trying again. That matters more than it sounds. Nothing is queued behind your back, so you never discover that a backlog emptied itself overnight. A message you type yourself in the conversation goes out right away and skips this mechanism, because there's no reason to slow it down.
The caps aren't the same from one network to another, and LinkedIn's is deliberately the lowest of the three (the network's own limits are detailed in LinkedIn message limits). A restriction there costs you a professional account, which justifies a wider margin. These values aren't preferences you adjust. They're limits the product holds.
There's no time-of-day restriction, though. A follow-up you approve late at night goes out late at night, because refusing to send while you're sitting in front of your screen would be friction with nothing to show for it. Quiet hours don't block any message: they only shift the time your morning follow-ups get prepared.
Finally, a freshly connected account imports its history in successive batches rather than in one block, so it works at a natural pace from day one.
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. For today's follow-ups, that sentence is literal: each message waits for a click. Ficiel also has automations that can send without a click per message, but they only exist if you set them up and switch them on yourself, and you write the text in advance.
What we don't promise
No tool can guarantee that Instagram will never ask you for a security check. That's actually what we tell users when they connect an account: in the first days, the platform notices a login from a new device and may send a security message. It's expected, and there's nothing to do except confirm it was you.
That honesty has practical value. A user who was warned confirms and moves on. A user who was promised nothing would ever happen gets scared, ignores the alert, and ends up losing access to the channel. The reassuring promise costs more than the warning.
We don't promise a numeric result either. What the product holds is a way of sending. What you get out of it depends on your history: how many leads went quiet, how warm they were, how long ago they wrote. The honest way to estimate it is to look at your own inbox, not at a sales page. That's also why the follow-ups worth sending are the ones in old conversations you already had, not new ones you'd have to start.
Two questions to ask before choosing a tool
The first: does it write to people who never talked to me? If yes, pacing becomes secondary, because the recipient carries most of the risk. A tool that goes after strangers from a list exposes you no matter how careful it is technically.
The second: can I read the message before it goes out? The answer splits tools into two families. Those that show you the message and wait, and those that show you a report afterward. The first family lets you catch a mistake of context or recipient. The second lets you watch it happen.
These two questions beat any feature comparison, because they're about how the tool is built rather than which options it has. If you want to go further on the second one, we wrote about why keeping a human in the loop is the safeguard that catches everything else.
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. On the question of pacing, that means it only writes to people who are already in a conversation with you, which is the least exposed way to send.
How do I follow up with leads on Instagram without getting banned?
Only write to people who are already in a conversation with you, space your messages irregularly, and make each message different. That's the rhythm Instagram looks at, not a published number. No tool can guarantee Instagram will never ask for a verification. Ficiel only follows up on existing conversations, and you approve each follow-up before it goes out.
How many DMs can I send per day on Instagram?
There's no public limit, and any content that gives you a precise one made it up. These limits depend on the age of the account, its history and its recent behavior. The right move isn't to aim for a number. It's to spread your messages out, vary them, and stay in a volume comparable to what a person would handle by hand.
Should I wait before sending from an account I just connected?
Yes. The first days are the most fragile period, because the platform sees a new session. Ramp up gradually rather than sending a lot right away, and above all limit bulk reading of profiles or posts, which weighs more than a message does.
Is regular spacing between messages enough to look human?
No, and it's a common mistake. A perfectly regular cadence is a recognizable pattern in itself. What looks human is an interval that varies from one message to the next, combined with messages that don't look alike.
Do quiet hours stop a message from going out?
No. In Ficiel, a message you approve goes out when you approve it, day or night. Quiet hours only shift the time your morning follow-ups get prepared. What protects the account is the caps and the spacing, which apply all the time.
