How-To Guide
What are Instagram auto-replies, and what can they do?
What Instagram's own auto-replies are, which message each native feature covers, the rules every tool inherits from Meta, and how to tell whether yours works.
By the Superdirector editorial team. Meta's messaging policy, Send API and platform overview pages, plus six practitioner guides, were read in full on November 14, 2026. Meta's Business Help Center and Instagram Help Center pages were not directly readable, so in-app menu names carry the practitioner sources that quote them.
At a glance
What are Instagram auto-replies, and what can they do?
An Instagram auto-reply is a message that sends itself when something you defined happens: a first direct message, a message outside your hours, a tap on a published question, or a keyword in a message or a comment. Meta's own versions are free, and each one covers a single moment. Most of them read a word, not a sentence, so the message you write has to work without the question it is answering. The same rules apply to every tool in the field, free or paid: the person starts the conversation, you answer inside the window their action opens, and a comment earns one private reply. What the feature cannot do is what decides whether you need it at all.
- Five native features cover five moments: a first-message greeting, away hours, tapped questions, keyword replies, and comment to message.
- Meta's own replies are free. Paid tools change what the message can carry, not the rules that allow it to be sent.
- Native replies match exact words, so a question worded twenty different ways reaches you untouched unless a person answers it.
- The hard limits belong to the platform, not the vendor: no cold messages, no messages to new followers, and no reading of typed text unless a tap or a keyword fired.
Five separate triggers, each matched to the one native feature Meta ships for it. Original diagram, hand-drawn from the feature table in this section read November 14, 2026.
Make it work for your brand
Find it. Remake it. Post it. Superdirector turns the viral formats in your niche into your brand's carousels, X posts and video scripts, and keeps the calendar full.
What an Instagram auto-reply actually is
An Instagram auto-reply is a message that sends itself when a condition you defined is met. Klaviyo's definition puts the trigger first: it is an automated response triggered by a specific action, which can be a comment with a keyword, a direct message or a story mention. That wording matters. The feature is not one button. It is a set of triggers, and each trigger decides who receives the message and when.
The phrase covers at least five different features, which is why people follow a tutorial and land in a settings screen that does not hold the thing they wanted. Pinlyx's guide lists them in one place: saved replies, chat questions, instant reply, away message, and the keyword and comment rules. Four of those send on their own. One does not.
Saved replies are the one that waits for you. You type a shortcut and the saved text appears in the message box, then you send it. Nothing leaves the account until a person taps send. It saves typing, and it is not automation.
Every one of these routes also needs a professional account. Blotato's guide to the three paths says it plainly: none of them work on a personal account. Switching to a business or creator account is free and reversible, and it comes before any other setup step.
This page covers what each feature sends and what none of them can do, because that is the decision a buyer makes before choosing a tool. The click-by-click setup lives on the setup walkthrough, and the numeric ceilings live on the Instagram DM limits page.
The five moments a reply can answer
Meta ships ten templates inside Meta Business Suite, but five of them carry the actual reply work. Each one answers a different moment, and each one stops in a different place.
| The moment | The native feature | What it sends | Where it stops |
|---|---|---|---|
| Someone messages the account for the first time | Greeting, called Auto reply or Instant reply | One message, the same for everyone | It does not read what the person wrote |
| A message arrives outside your hours | Away message | Your off-hours note | It fires on a schedule, not on the question |
| Someone opens a chat and taps a published question | Frequently asked questions in the Instagram app | The answer to that question | Only on a tap, and only up to four questions |
| A keyword arrives in a direct message | Custom keywords in Business Suite | The reply mapped to that word | Exact and case-sensitive matching, up to five keywords, desktop setup |
| A keyword appears in a comment on a post | Comment to message in Business Suite | One private reply to the commenter | Desktop setup, and one private reply per comment |
The keyword rules deserve their own sentence, because they explain most of the disappointment. Keyword matching is exact, and capitalization counts. Only five fit in an automation, and a keyword can belong to one automation. SocialMaker's guide also records the delay: the reply is not instant, because Meta sends it after 15 minutes unless a person answers first. Real questions, typed the way people type, will often miss all of that.
Meta's creator AI replies are a sixth option and a different animal. Instead of a fixed message, a creator AI answers in your style from a persona you built, and you choose the audience it answers. Blotato's read of the feature notes the two limits that stop most readers: it needs a creator AI you have already built, and Meta says it is being gradually introduced and may not be available to you yet. It also works only in the mobile app.
One naming warning. Meta's desktop template is called Auto reply, while some of Meta's own help pages call the same thing Instant reply. They are the same greeting. If a guide tells you to look for one and you see the other, you are in the right place.
What every auto-reply inherits from Meta
The rules below come from Meta's developer documentation, and they apply to every tool that connects through the official API. No vendor can opt out of them, and a vendor who claims to is telling you something useful about how the tool works.
Who may speak comes first. Meta's platform overview states that a person must initiate the conversation, and the Send API page repeats that the conversation must be initiated by that person. An auto-reply is always a response. There is no compliant way to open a thread with someone who has not messaged, commented or replied to a story first.
How long the reply may go out is the next constraint. The standard messaging window depends on the action that opened it, and for most actions it lasts 24 hours. A first message after a Click-to-Messenger or Click-to-Direct ad can open a window of up to 7 days. Outside that window, the Human Agent tag lets a business answer manually within a separate 7-day period.
Comments carry their own clock. A private reply is a single message, and it must be sent within seven days of the person publishing the comment. One comment, one private reply, one chance to carry the link or the code.
Anything that behaves like a bot has to meet two tests. Automated bots must respond to any and all input from the user and answer within 30 seconds, because Meta treats a slower or partial answer as a broken conversation. Where the law requires it, an automated chat must also disclose that the person is talking to an automated service, at the start of the thread, after a significant lapse of time, or when a human hands the chat back to automation.
Two more rules shape what the conversation can become. A person can block or mute a business conversation at any time, and messages sent inside the standard window may contain promotional content. High-volume messaging can also temporarily disable inbox functionality, which is worth knowing before a viral post turns one automation into a flood.
Read together, these rules explain why every serious tool looks similar at the edges. The differences sit in what the message can carry and what happens after it lands, not in who is allowed to send.
What an auto-reply cannot do
Five jobs sit outside every official path. Knowing them saves a subscription, and it also tells you when a tool's marketing is describing something the platform forbids.
- Message a stranger first. The person must initiate, so cold DMs are not an official feature of any tool.
- Message every new follower. A follow does not open a conversation, so there is no thread to reply in.
- Read a typed question with the native tools. Tapped questions answer a tap, and keywords answer a word.
- Keep a customer record, route by intent, or hand the chat to a teammate. Meta's native tools do none of these, which is the honest gap a third-party tool fills.
- Follow up when someone does not tap, reply or click. The native reply ends the interaction; the next move belongs to the person.
That list is the real buying question. If your inbox holds the same handful of questions, the free features cover it. If it holds questions worded a hundred ways, a keyword will miss most of them, and no amount of template writing fixes a match rule that reads one word.
One more thing that looks like a gap and is not: an auto-reply cannot make a slow business fast. A greeting that promises an answer within a day is a promise the inbox has to keep. Writing a reply time you cannot meet turns the feature into a source of complaints.
When the free tools are enough
Begin with the shape of the messages, not the size of the account. Three shapes cover most cases. One is a small set of repeat questions: hours, shipping, size, price. Another is many different questions, each worded in its own way. The third is a specific post or reel that asks people to comment for something.
The first shape is where the free tools earn their place. A greeting that says when you will answer, an away message that covers the weekend, and four tapped questions for the things you answer daily will handle a good share of a small inbox. They cost nothing and they set an expectation.
The second shape is where keywords start to fail. Match rules that need an exact word will not catch a sentence. That is the point to consider a tool connected through Meta's official login, because it can read the message instead of a token inside it.
The third shape is a campaign decision, not an inbox decision. If a post asks people to comment a word, the comment template can carry one private reply, and it has to go out inside seven days. A tool becomes worth its price when the reply needs a link button, a saved contact, a follow-up, or a hand-off to a person.
Two switches surprise people after they connect anything. Turning on Meta Business Agent pauses the automations you already had, including away messages and instant replies. And only one inbox app receives incoming messages, so a third-party inbox and Meta Business Suite Inbox cannot both be primary.
The shape of your messages, not the size of the account, decides whether Meta's free replies are enough. Original diagram, hand-drawn from the three cases described in this section.
How to tell whether yours is working
Four readings tell you whether the feature is doing its job, and none of them needs a new dashboard. First, how long it takes a person to answer the messages the automation did not catch. Second, how many incoming messages matched an automation and how many did not. Third, what the person does after the auto-reply: replies, taps the link, books, or goes quiet. Fourth, whether anyone blocks or mutes the account after a greeting.
Each reading points at a different fix. A slow human reply means the greeting's promise is wrong, so change the promise or staff the inbox. A high share of unmatched messages means the keyword set is too narrow, so add the phrasings people actually use, up to the five the automation allows. A quiet reply after the greeting means the greeting asks for nothing, so give it one clear next step. A rise in blocks or mutes means the message repeats too often or sounds like a broadcast.
Keep one rule while you tune it: test from a second account. Meta's own guides note that a new automation can take a few minutes to appear and may need a refresh, so a first test that seems to fail may only be early. Test again after the template has settled.
Where Superdirector fits, and where it does not
Superdirector does not answer DMs. No plan sends an Instagram auto-reply, and saying so plainly is more useful than bending the feature list to fit the question.
What the plans do is separate. The $29 plan is MCP access plus hands-on creation: live social data, ideas, scripts and downloads inside Claude or ChatGPT, with no posting. A builder who wants their own DM automation can start there, wire it through Meta's API and keep ownership of the tool. The reply still follows Meta's rules, and the work is yours.
The $199 plan runs one AI-run operated account that publishes eligible Instagram carousels and single link-free X text posts by itself once posting is turned on, under account-level standing posting permission. That is the posting job, not the inbox job. It does not read or answer a direct message.
So the choice is a clean one. If the problem is what to publish, the operated account covers it. If the problem is what to answer, keep the free replies for the moments they cover and add a tool only for the messages that need more than a matched word.