We make PostWharf, a scheduler that gives every post a receipt with one line per network, says in plain words why a line failed, and emails the workspace's owners when a post does not go out. Plans start at $15 a month for one brand, with as many channels as you connect: see what each plan includes.
This page is for posts that never went out at all. If your tool said a post was published and it is not on the network, that is a different problem, covered in why scheduled posts fail silently.
How this list was put together
We started from the causes that the help-centre pages ranking for this question name, searched on 13 November 2026, and checked each against a network's own developer documentation or PostWharf's own code where one states the rule; causes 7, 9 and 10 rest on the help-centre pages alone. The network figures come from Meta's Instagram content publishing guide and Facebook token guide, LinkedIn's Posts API and authorization guide, TikTok's Content Posting API, Google's OAuth guide and Pinterest's authentication guide, read on 10 November 2026. The character and file limits are the ones PostWharf enforces for each network before a post is queued.
The ten causes at a glance
Find the message your tool shows in the second column. Where it shows only "failed" or "error", the last column tells you whether sending it again unchanged can work.
| # | Cause | What you will usually see | Fix | Send the same post again? |
|---|---|---|---|---|
| 1 | The account's sign-in ended | Every post to one account fails from a certain day; the tool asks you to reconnect | Reconnect the account | Yes, after reconnecting |
| 2 | The person who connected it lost a role or permission | Posts to a Page or channel fail while that person's own profile still works | Restore the role, or reconnect through someone who has it | Yes, once access is back |
| 3 | The post broke a network rule | "Too long", "unsupported media", "file too large", or a refusal naming the file | Shorten the text or change the file for that network | No, change it first |
| 4 | The network flagged it as spam | A refusal that names spam or risk, often with no fix offered | Rewrite the text; stop posting through that account for a while | No |
| 5 | A rate limit or daily cap | "Rate limited", "too many requests", or a daily-cap message | Wait for the window to reset, or space the posts out | Yes, after the wait |
| 6 | The time was wrong | The post went out at the wrong hour, or was refused for a time in the past | Check the time zone the tool schedules in; pick a future time | Yes, with a corrected time |
| 7 | It was never released | Still listed as a draft, awaiting approval, or waiting for you to finish it on your phone | Release it, approve it, or post it by hand | Yes |
| 8 | The tool's own plan or payment lapsed | Queued posts cancelled together on one day, with a note about the plan | Restart the plan or update the card | Yes, after that |
| 9 | The network or the tool did not answer | "Timed out", "service unavailable", or no answer at all | Wait; check the network's status | Yes, once it is back |
| 10 | It went out later, or somewhere else | Nothing failed, but the post was not where or when you expected | Check the account it went to and any timing offset | Usually not needed |
1. The account's sign-in ended
When you connect an account, the network gives the scheduling tool a sign-in that stands in for your password, and the network decides how long it lasts. On Instagram, Threads and LinkedIn it runs for 60 days unless the tool renews it, and Pinterest's access lasts 30 days on the same terms. Changing your password, removing the tool's access from the network's own settings, or a security check on the network's side can end it early.
How to tell: posts to one account stop working from a particular day, while every other account in the tool carries on. Most tools mark the account as needing reconnection.
Fix: reconnect the account in the tool, then send again each post that missed it. Reconnecting does not send them for you. Why each network ends sign-ins, and how far ahead you can see it coming, is in why social accounts disconnect from scheduling tools.
2. The person who connected it lost a role or permission
A post to a business Page goes out under the role of whoever connected it. If that person left the company, or was moved to a role that cannot post, the sign-in still works but the Page refuses the post. LinkedIn, for example, lets only an administrator, a content admin or a direct sponsored content poster publish to a company Page. On Telegram, the bot has to be an administrator of the channel with permission to post.
How to tell: posts to the Page fail while posts to that person's own profile still go out. The refusal usually mentions permission, access or a role.
Fix: give the role back, or have someone who holds the right role reconnect the account. Network-by-network detail for LinkedIn Pages is in LinkedIn scheduled posts not publishing.
3. The post broke a network rule
Each network sets its own limits on text and files, and they differ widely. X takes 280 characters and counts every link as 23 of them, Bluesky takes 300, Threads 500, Instagram and TikTok 2,200 for a caption, and LinkedIn 3,000. Files have their own rules: X takes up to four images or one video on a post, not both, and a YouTube post is a video, so it needs one. A post written once and sent everywhere often fits most networks and breaks one.
How to tell: the post went out to some networks and failed on one, with a message about length, format, size or the number of files.
Fix: write a shorter version or pick a different file for that one network, then send it again. Sending it unchanged fails the same way. Instagram's own wording for each refusal is decoded in Instagram API error codes.
4. The network flagged it as spam
Some networks refuse a post they judge risky. TikTok documents this openly: its refusals include one for text it rates as spam, one for a creator who has posted too much through apps that day, and one for an account barred from posting, and for some of these it tells apps outright not to try again. An app TikTok has not yet audited can post only to private accounts.
How to tell: the refusal names spam or risk, or the account was posting heavily just before.
Fix: change the text, slow down, and do not resend the same post straight away. A post refused as spam will be refused again.
5. A rate limit or daily cap
Networks cap how much any one app or account can publish. Meta's content publishing guide puts Instagram accounts at 100 posts published through the API in a rolling 24 hours, although another section of the same guide gives 50. TikTok allows at most five uploads waiting for a creator to finish within any 24 hours. A busy day, or several tools posting to one account, can reach these.
How to tell: the message says rate limited, too many requests or a daily limit, often for several posts in a row.
Fix: wait for the window to pass and send again. A good tool tries again on its own after a rate limit, rather than marking the post failed straight away.
6. The time was wrong
The most common version is a time zone: the tool scheduled in one zone and you meant another, so the post went out hours early or late. The other is a time that had already passed when the post was queued. Some tools send that post immediately, others refuse it.
How to tell: the post did go out, at the wrong hour, or the tool refused it for a time in the past.
Fix: check which zone the tool schedules in, per brand or per account if it has more than one, and pick a time that is still ahead of you.
7. It was never released
A post can sit in a tool and never go anywhere. It may still be a draft. In a team, it may be waiting for an approval that nobody gave before its time. And some tools have a reminder mode that sends the post to your phone instead of publishing it, so it goes out only when you finish it by hand.
How to tell: the post is still listed as a draft, pending or awaiting approval, or a reminder arrived that nobody acted on.
Fix: release or approve it, or post it by hand from the reminder.
8. The tool's own plan or payment lapsed
When a subscription ends or a payment fails, a scheduler usually stops sending. Posts queued for after that day are held or cancelled, all at once and on every network.
How to tell: every post due after one date stopped together, and the tool shows a note about the plan or billing.
Fix: restart the plan or update the card, then send the posts again. In PostWharf a post stopped this way says so on its receipt, names what to change in Settings, and was never sent to any network.
9. The network or the tool did not answer
Networks have outages and slow spells, and so do scheduling tools. A post due in that window gets a timeout or no answer at all.
How to tell: a timeout or "unavailable" message, often on many accounts at once, or the network's status page reports a problem.
Fix: wait until the network is back, then send it again if your tool has not already done so. Check first whether it went out after all, so you do not post it twice.
10. It went out later, or somewhere else
Sometimes nothing failed. A few schedulers offer a setting that moves posts a few minutes either side of the chosen time so they look less automated. A post that was retried after a rate limit goes out later than planned. And a post sent to the wrong account or Page looks, from the right one, as if it never went out.
How to tell: the tool shows the post as sent, but at a different time or to a different account than you expected.
Fix: check the account the post went to and any timing setting. If the tool says it went out and it is nowhere on the network, see the silent-failure guide linked at the top of this page.
What to check on each network
Most failures look the same everywhere, but each network has one or two things worth checking first.
- Instagram: a 60-day sign-in, the daily publishing cap, and a 2,200-character caption.
- Facebook: a Page connection has no end date, but it can stop when the person who made it changes their Facebook password or removes the app.
- LinkedIn: a 60-day sign-in, and the connecting person's role on the company Page.
- X: 280 characters with links counted as 23, and four images or one video, never both.
- TikTok: the daily cap on posts made through apps, spam refusals, and private-only posting for an unaudited app.
- YouTube: every post needs a video, and Google cancels a sign-in that has gone unused for six months.
- Threads: a 60-day sign-in and a 500-character limit.
- Pinterest: a 30-day access period, after which the account needs reconnecting.
- Bluesky: the app password the tool uses can be revoked on its own, and a post holds 300 characters.
- Telegram: the bot must be a channel administrator, and a caption on a photo or video is limited to 1,024 characters against 4,096 for a text message.
PostWharf does not publish to Reddit, Discord, Snapchat or WordPress yet, so they cannot be connected or picked for a post.
How to find out it failed before your audience does
The worst way to learn a post failed is from a customer asking why the launch went quiet. Four habits catch it sooner.
- Turn on failure alerts, and send them to someone who reads them. An email to a shared inbox nobody watches is the same as no email.
- Read the status per network, not per post. A post that reached four networks and missed the fifth can look fine at a glance.
- Check connections a week ahead. Any account whose sign-in ends before your next important post needs reconnecting now, not on the day.
- For posts that matter, look at the network itself. A tool's "sent" is its record of what it did, and the network's page is the proof.
What to look for in a scheduler's alerts, and how to tell a delay from a real failure, is set out in social media post failure notifications.
What PostWharf does when a post cannot go out
PostWharf catches several of these causes before the post is queued, rather than at its scheduled time. It refuses a time that has already passed, and through the API the refusal shows the exact time it read, so a time-zone mix-up is plain to see. It refuses text that is too long for a network, measured the way that network counts, and files a network will not take, naming the network and the limit. And it will not queue a post for a channel that is added but not connected. The composer also warns you before sending to an account whose sign-in has ended.
For everything else, the receipt does the work. Each post has one line per network with its status, the link once it is live, and the reason in plain words when it failed, along with the fix: reconnect, shorten, change the file, or wait. When a network is briefly unavailable or asks it to slow down, PostWharf tries again on its own. When a network goes quiet in the middle of a publish, PostWharf does not send it again blind: the receipt asks you to check the network first, because the post may already be there. A post that needs a change goes straight to failed with the reason, rather than being retried into the same refusal.
When a post does not reach a network, every owner of the workspace gets an email naming the post and the networks it missed, with several failures from one pass gathered into a single message. When a sign-in stops working, or will run out within the week, the owners get an email naming the account, the posts that already missed it, and how many more are queued for it. These emails are on by default and can be switched off in Settings. The receipt's resend button, named for the networks that missed ("Send Instagram again"), re-sends the post only to those networks, so the ones that already have it do not get a duplicate, and you can give one network different files before you do. The same events go to webhooks and the API, for teams that want failures in their own alerting. Creating an account takes no card, and Starter's first seven days are free: compare the plans.
Common questions
Can I just reschedule a failed post?
Only when the cause was a rate limit or an outage, and a post that simply went out late needs nothing at all. For the other seven causes, the same post at a new time fails the same way. Reconnect the account, restore the role, or change the text or file first.
Why did my post publish on some networks but not others?
Each network checks the post against its own rules and its own sign-in, so one post can pass four networks and fail on the fifth. The usual reasons are a length or file limit that only one network has, or a sign-in that has ended on that one account. Send it again to the network that missed it, not to all of them, or the others get it twice.
How do I stop this happening again?
Use a tool that tells you when a post fails and when a connection is about to end, check connections a week before important posts, and write a separate version for any network whose limits the main version breaks. If you want that done for you across every network a brand posts to, PostWharf plans start at $15 a month.
PostWharf is priced per workspace rather than per connected channel, and every plan carries unlimited channels. See per-brand pricing.