Why "sent" does not mean read
Instagram will tell you it accepted your message. It will never tell you anybody saw it. That gap is where most automation dashboards quietly flatter their owners, and here is what to measure instead.
Here is a number you have probably seen on a dashboard. 1,284 messages delivered. Next to it, maybe, an open rate. Ninety something percent, because DM open rates are famously enormous.
Almost none of that is measured. Most of it cannot be.
This is not an accusation against any particular company. It is a structural fact about what Instagram tells you, and once you see it you cannot unsee it on a pricing page.
The short answer
Instagram tells you it accepted your message, and nothing after that. There is no delivery receipt, no read receipt, and no endpoint you can call tomorrow to ask what became of it. Any dashboard showing an open rate for Instagram DMs has derived that number from something else, usually link clicks or an industry average.
What Instagram actually gives back
When a private reply succeeds, Meta’s documented response contains two things, an id for the recipient and an id for the message.
That is the whole of it. There is no delivery receipt. There is no read receipt. There is no endpoint you can call tomorrow to ask what became of message id such and such. The API has told you one thing and one thing only. it took the message from you.
If that is surprising, it is worth seeing where in the sequence the message actually leaves. There is exactly one moment when anything goes to Instagram, and everything you can know comes from what it says back at that moment.
The wall in the middle
Between Instagram accepting your message and a human being reading it sits the recipient’s own settings, and they are not a formality.
If somebody does not follow you, your message is a message request. It sits in a separate place that many people check rarely and some people never check at all. Some accounts turn off messages from strangers altogether. None of this is reported back to the sender.
So when a tool shows you “delivered”, it has made an assumption about everything in that figure after the first box. Usually that assumption is simply that acceptance equals delivery, which is a reasonable shorthand right up until you use it to decide where to spend your month.
Open rates are worse, because nothing in the platform exposes an open. An open rate on a comment to DM dashboard is either counting clicks on a tracked link and calling them opens, or it is an industry average presented as your number.
Why anybody does this
Not, mostly, because they are lying. Three ordinary pressures produce the same result.
The first is that “sent” is a weak sounding word and every competitor is using a stronger one. The second is that a tracked link click genuinely is evidence of something, and it is tempting to promote it into evidence of more. The third is the deepest. A dashboard with an honest gap on it looks broken. Customers ask why you cannot show them what everyone else shows them, and the cheapest answer is to show it.
We decided not to, and wrote the decision down where it is hard to quietly reverse. Our own measurement contract has one line in it that governs this.
Accepted by Instagram means Instagram took the message from us. Not read, not converted. Never reported as either.
It is a rule written against our own future temptation, which is the only kind of rule worth writing.
What to measure instead
The useful thing about giving up a fake number is that it forces you to find a real one. Three are available to anybody, and all three are better than an open rate.
Accepted, stated plainly. Keep it, and keep calling it what it is. It is a genuine measure of whether your automation is working, which is a real question with a real answer. It is simply not a measure of whether anybody read anything.
Clicks on the link you sent. If your message carries a link, a click is a human action that actually happened. It is the first point in the whole chain where somebody did something. This is a far more honest headline than an open rate, and it is usually a far smaller number, which is the point. A small true number tells you where to improve. A large false one tells you nothing.
What happened afterwards. The reply you got back, the sale, the booking, the person who turned up. This is the only number that pays for anything, and it is the one most dashboards are worst at, because it happens outside the tool.
All three are downstream of the message itself, which you only get one of. What goes in it is a smaller question than the industry makes it and a more constrained one.
The follower asymmetry nobody mentions
There is a practical consequence of the message request wall that is worth building around.
Your message is far more likely to be seen by somebody who already follows you, because it goes to their actual inbox. Which means the same automation on the same post performs very differently depending on who is commenting, and the difference is invisible in your “delivered” figure.
If a campaign aimed at strangers underperforms one aimed at your own audience, that is not necessarily your copy. It may be that a chunk of the strangers never saw the message at all. Judging those two campaigns by the same number will teach you the wrong lesson.
How to read a pricing page after this
Three questions, and you can ask all of them in an email.
- What exactly does this number count? If “delivered” means “the API accepted it”, that is fine. It just needs saying.
- Where does the open rate come from? There is no open signal in the platform. Any open rate is derived from something else, and you are entitled to know what.
- What happens when a message fails? Do you see Instagram’s actual reason, or a shrug? This matters more than it sounds.
A tool that answers all three clearly is probably being straight with you elsewhere too. A tool that cannot answer the first one is showing you a number it has not thought about, which is its own kind of answer.
Where this came from
- Meta, Private Replies, Instagram Messaging documentation Read 29 October 2026. The documented success response returns a recipient id and a message id, and nothing about delivery or reading.
Try it on your own account
Free to open, and you can see exactly what it would send before any of it reaches Instagram. See how it works.
Create a free accountWorks with Instagram Business and Creator accounts, connected through Meta’s official integration.
Comments
Nobody has written yet. Yours would be the first.