Can Instagram ban you for automated DMs?
Yes, if the automation breaks Instagram's rules: a tool that logs in as you without permission, cold commercial messages, or ignoring people who say stop. Replies through Meta's official API to people who contacted you first are allowed, though messaging can still be limited.
Where common DM automations sit against the rules
Automation on its own is not the problem. Meta built an API for it, and our done-for-you Instagram DM automation runs on that API, through GoHighLevel, ManyChat or n8n. What gets accounts restricted is how the automation behaves: who it messages, how it connects, and what it does when people object. This page lists each rule in Meta's own words, read October 30, 2026, and the behaviour that breaks it.
Meta's documents use the words "remove", "disable", "restrict", "feature limited", "paused" and "rate-limited". They do not publish a list of automations that lead to a permanent ban, so nobody outside Meta can promise which action gets which penalty. What you can control is staying inside the written rules.
What do Instagram's rules say about automation?
Instagram's Terms of Use forbid accessing Instagram "in an automated way without our express permission", its Community Guidelines forbid repeatedly contacting people for commercial purposes without consent, and Meta's developer policy requires opt-in and immediate opt-out for Instagram messaging.
We read the five Meta documents that govern automated Instagram DMs and pulled out every line that applies. The quotes are exact.
| Document | What it says | What it means for DM automation |
|---|---|---|
| Instagram Terms of Use, section 4.2 | You can't "access or collect information in unauthorised ways. This includes creating accounts or accessing or collecting information in an automated way without our express permission" | Tools that log in with your password or drive the app like a person are outside the official route |
| Instagram Terms of Use, section 4.2 | You can't "circumvent, bypass or override any technological measures that control or limit access to the Service or data" | Tricks to get past limits, such as rotating accounts to keep sending, are themselves a breach |
| Instagram Community Guidelines | No "posting repetitive comments or content, or repeatedly contacting people for commercial purposes without their consent" | Mass DMs to people who never asked, and copy-paste comment blasts |
| Meta Developer Policies 5.2(a) | "Don't contact people in Messenger or Instagram Messaging unless you" have "legally sufficient consent, permission, and/or opt-in" | The person must start the contact: a DM, comment, story reply or message-ad tap |
| Meta Developer Policies 5.2(b) | "Immediately respect all requests (on or off Messenger and Instagram Messaging) by people to block" or opt out | A "stop" must end the automation at once |
| Meta Developer Policies 5.4 and 5.6(1)(b) | Experiences with excessive negative feedback "may be feature limited, paused, rate-limited or removed"; "Monitor your block rate" | Even allowed flows get limited if people block or report them |
| Messenger Platform and IG Messaging API policy | "Use of tags outside of approved use cases may result in restrictions on your ability to send messages" | Message tags such as Human Agent are not a way to send promotions after the window closes |
The Community Guidelines also state the consequence: "Overstepping these boundaries may result in deleted content, disabled accounts or other restrictions."
Which kinds of Instagram DM automation carry the most risk?
Highest risk: tools that log in with your password, cold or mass DMs, and welcome DMs to every new follower. Lower risk: replies through Meta's API to people who messaged or commented first, with a working opt-out. Risk rises again with high block rates.
This matrix is our reading of the rules in the table above, applied to the automations people ask us about. "Risk" is our judgement from the wording; Meta does not publish penalty rates, and we have left out the ban-rate percentages that circulate on vendor blogs because none of them trace back to Meta data.
| What the automation does | Rule it touches | Our risk rating |
|---|---|---|
| Logs in with your username and password, or runs as a browser extension that clicks and types for you | Terms 4.2: automated access without express permission | High |
| Sends DMs to a scraped list, hashtag users or a competitor's followers | Developer Policies 5.2(a); Community Guidelines on commercial contact without consent | High |
| Sends a welcome DM to every new follower | A follow does not open a messaging window in Meta's public API, so outside Meta's own gated Follow to DM beta (offered through ManyChat to accounts Meta selects) the tool must work outside it (Terms 4.2) | High |
| Keeps sending promotions after the 24-hour window by using message tags | Messenger and IG messaging policy on tags | High |
| Ignores "stop" or keeps messaging someone who blocked the flow | Developer Policies 5.2(b) | High |
| Allowed triggers, but a pushy flow that many people block or report | Developer Policies 5.4 and 5.6 | Medium |
| One private reply to a keyword comment, through the API | Allowed: Meta's Private Replies docs permit one message per comment within 7 days | Low |
| Answers DMs and story replies inside the 24-hour window, with a stop path and a human handover | Allowed: Meta's messaging docs and Developer Policies 5.2 | Low |
The pattern is simple: anything that starts a conversation the other person did not start, or that connects outside Meta's API, sits at the top. The official API prevents most of it by design. Meta's messaging docs say "Conversations only begin when an Instagram user sends a message to your app user", and private replies allow "only one message" per comment.
Is ManyChat or GoHighLevel safe to use for Instagram DMs?
A tool that connects through Meta's official login, where you approve permissions instead of handing over your Instagram password, works inside the API's limits. That makes the tool itself low risk. Your flows can still get the account limited if they spam.
The quickest test for any tool: how did you connect it? HighLevel connects through Facebook's authorization window, where you pick your Pages and approve permissions, and its help articles follow Meta's 24-hour messaging rule. A tool that asks for your Instagram password, or needs a browser left open on your computer, is acting as you, which is the automated access Instagram's Terms prohibit without permission.
Searches for "is ManyChat safe for Instagram" and "can ManyChat get you banned" usually mean the second part: can an approved tool still get an account restricted? Yes, through what it sends. Meta's Developer Policies say messaging experiences with excessive negative feedback "may be feature limited, paused, rate-limited or removed", whichever tool sends them. Our GoHighLevel comment-to-DM setup guide shows a build that stays inside the rules.
What happens if Meta limits your Instagram messaging, and what should you do?
Meta's documents describe limits that range from rate limiting and pausing a messaging experience to removing content and disabling accounts. If sending is limited, pause the automation, find which rule it broke, fix that, and restart slowly with lower volume.
What to do when messaging is limited
The steps we follow if an account's messaging is limited:
- Pause every automated send. Keep answering people by hand inside the normal window.
- Check how each tool connects. Disconnect anything that uses your password or a browser extension.
- Check who the flows message. Every flow should start from a DM, comment, story reply, story mention or message-ad tap.
- Check opt-outs. A "stop", "unsubscribe" or block must end the flow at once (Developer Policies 5.2(b)).
- Look for negative feedback. Which flow gets the most blocks, reports or angry replies? Rewrite or remove it.
- Restart one flow at a time, lowest-risk first, and watch replies for a few days before turning on the next.
Busy reels can also look like a limit when they are not. Meta caps private replies to post and reel comments at 750 calls per hour per account, so on a viral post some DMs arrive hours late while nothing is wrong with the account.
How do you keep automated Instagram DMs inside the rules?
Connect only through Meta's official API, message only people who contacted you first, send one private reply per comment, stop at once when someone opts out, say the assistant is automated, and hand anything sensitive to a person.
Our pre-launch checklist, with the rule behind each item:
| Check before launch | Rule behind it |
|---|---|
| Every tool connected through Meta login and permissions; no passwords shared | Instagram Terms 4.2 |
| Every flow starts from an allowed trigger | Developer Policies 5.2(a); Meta messaging docs |
| One private reply per comment, ending in a question or buttons | Meta Private Replies |
| "Stop" ends the flow; a block is respected | Developer Policies 5.2(b) |
| The first AI message says it is an automated assistant | Messenger and IG messaging policy on disclosure |
| No message tags used for promotions | Messenger and IG messaging policy on tags |
| A named person reviews blocks and complaints weekly | Developer Policies 5.4 and 5.6(1)(b) |
The AI side follows the same rules as our AI chat agent service: approved content only, a fallback for every input and a clear handover. Our article on what GoHighLevel automation does in practice covers why a flow that annoys people costs more than it earns.
In the systems we build, consent and opt-outs are logged in the CRM next to each lead, the same way our lead automation handles every other channel, and you keep ownership of the automations and data.
More questions about Instagram DM automation and bans
Can Instagram ban you for sending DMs by hand?
Yes, the same Community Guidelines apply to people as to tools. They ask users not to repeatedly contact people for commercial purposes without their consent, and they say breaking the guidelines may result in deleted content, disabled accounts or other restrictions. A sales team copying the same pitch into hundreds of cold DMs is the manual version of a spam bot.
Is ManyChat against Instagram's terms of service?
Instagram's Terms forbid accessing Instagram in an automated way without Meta's express permission. Tools that connect through Meta's official messaging API, where you log in with Meta and approve permissions, are using the permission route Meta built for businesses. What can still breach the rules is the content and targeting of your flows, such as messaging people who never contacted you.
Will Instagram ban me for auto-DMing new followers?
Outside one exception, it is a high-risk automation. Meta's public messaging API does not open a conversation when someone follows you, so a tool that welcomes every follower usually works outside the API, which is the automated access Instagram's Terms prohibit without permission. The exception is Meta's gated Follow to DM beta in ManyChat, limited to accounts Meta selects and one message per person per week. Otherwise, invite new followers to comment a keyword or reply to a story.
How many DMs can you send a day without getting banned?
Meta does not publish a safe daily DM number for accounts, and numbers quoted on blogs are guesses. Its published API limits are 750 private replies per hour for post and reel comments and 100 Send API calls per second for text, per professional account. Those are technical caps, not permission: one unwanted message can still be reported.
Does saying "this is an automated assistant" lower the risk?
It meets Meta's disclosure rule where the law requires it, and Meta recommends it even where the law does not. Its policy says automated chat experiences should disclose that a person is talking to an automated service at the start of a conversation, after a long gap, or when a chat moves from a person to automation. It does not make cold messaging allowed.
What should you fix before asking Instagram to review a restriction?
Remove the cause first. Disconnect any tool that uses your password or a browser extension, stop messaging people who never contacted you, make sure a stop request ends every flow, and rewrite flows that draw blocks. Otherwise the same restriction can return after a review. If you want a second pair of eyes on your setup, our free strategy call covers it.
Sources and further reading
- Instagram Terms of Use (section 4.2 on automated access, section 6 on enforcement), read October 30, 2026.
- Instagram Community Guidelines (spam and consequences), read October 30, 2026.
- Meta Platform Terms and Developer Policies (sections 5.2, 5.4 and 5.6), read October 30, 2026.
- Meta: Messenger Platform and IG Messaging API policy (disclosure, message tags), read October 30, 2026.
- Meta: Instagram Messaging API (when conversations begin, messaging windows), read October 30, 2026.
- Meta: Instagram Platform, Private Replies, read October 30, 2026.
- Meta: Instagram Platform overview (rate limits), read October 30, 2026.
- HighLevel: Connect your Facebook Page with Unified Facebook Connect, read October 30, 2026.
