On the morning of June 12, 2026, Meta had one of its worst outages in months. Starting around 9:30 a.m. ET, Facebook abruptly logged people out and would not let them back in, Instagram feeds stopped loading, and Downdetector logged well over 100,000 reports inside the first half hour. It took roughly four hours to recover, region by region.

We build DM automation for a living, so the messages started arriving fast: is my automation broken? Did I lose leads? Is it something I did? Here is the honest version, including the part most tools will not tell you.

What Actually Happened on June 12

This was not a niche glitch. The symptoms (people getting logged out and unable to sign back in with the correct password, authentication errors rather than network errors) line up with a problem in Meta's backend login infrastructure, similar to the March and December 2024 outages. Meta did not publish a specific cause, which is normal for these events.

What matters for anyone running a business on Meta is which systems went down. It was not just the consumer apps. Meta's own business status page flagged high disruptions across the tools businesses actually depend on:

  • Facebook Ads Manager and the Marketing API, with a note that some advertisers could not create or edit ads
  • Meta Business Suite
  • Messenger Platform and the Messenger API for Instagram
  • WhatsApp Business Platform
  • Graph API, the core interface every messaging tool talks to, at medium disruption

Read that list again, because it is the important part. Those are the exact pipes every DM automation tool runs on. ChatGenius, ManyChat, the new Meta Business Agent, a custom script someone built in house: all of them send and receive messages through the Messenger Platform and the Graph API. When those go down, nobody is sending an Instagram DM. Not us, not anyone.

The Honest Truth: No Tool Is Immune

You will see automation companies imply their platform "stayed up" during a Meta outage. Be skeptical of that. A messaging tool is a layer on top of Meta. If Meta's Send API is rejecting calls, your tool physically cannot push a message into Instagram, no matter how good its uptime is. Anyone claiming their bot kept delivering DMs straight through a Graph API outage is either confused about what went down or hoping you are.

So if everyone is equally stuck while Meta is down, does the tool even matter during an outage? Yes, and this is the real difference. It is not about whether messages get blocked for a few hours. It is about what happens to those messages when the blockage clears.

Lost vs. Delayed: The Only Question That Matters

When a customer DMs you during a Meta outage, one of two things is true once the platform recovers:

  1. The message is gone. The trigger fired into a void, the send failed once, nothing retried, and that lead never got a reply. The customer assumes you ignored them.
  2. The message is queued. The reply that could not go out was held, not dropped, and it sends automatically the moment Meta is back. The lead gets their answer a bit late instead of never.

That difference is entirely down to how the tool is built, and it is the thing worth asking any platform you pay for. A naive setup does a single fire-and-forget API call: if it fails, the message evaporates. A resilient one treats every send as something that must eventually complete.

How ChatGenius Is Built to Handle This

We have lived through enough Meta hiccups, including the long-running Instagram webhook instability that has come and gone since late 2025, to design for them rather than pretend they will not happen. A few concrete things that matter on a day like today:

Outbound replies queue instead of vanishing

ChatGenius does not fire a single message and forget it. Sends move through a database-backed queue with atomic claiming, so a reply that cannot reach Meta during an outage stays in the queue rather than disappearing. When the Send API starts accepting calls again, the queue drains and those replies go out.

A safety-net job catches anything stuck

On top of the main queue, a background job runs every 15 minutes specifically to find work that got stuck (for example, a batch that was mid-send when Meta returned an error) and recover it. Outages are exactly the kind of mess that leaves things half-finished, and this is the broom that sweeps up after them.

Rate-limit awareness so recovery does not make it worse

When Meta comes back after an outage, everyone's queued traffic floods in at once and Meta tightens its rate limits. ChatGenius tracks outbound volume per account and respects those limits, so your account is draining its backlog steadily instead of getting throttled or flagged for hammering the API the second it recovers.

Inbound messages are protected by Meta itself

Here is a reassuring detail most people do not know: when Meta cannot deliver a webhook to your tool (an incoming DM or comment), Meta retries that delivery for hours on its side. So the messages your customers sent during the outage are not lost in the ether. Once the connection is healthy, Meta re-sends them, ChatGenius receives them, and your automation responds. Late, but handled.

Put together, the goal is simple: on an outage day, your worst case should be "replies went out a couple hours late," not "I lost a morning of leads."

What You Should Actually Do During a Meta Outage

If you woke up to today's outage and panicked, here is the calm checklist for next time:

  • Confirm it is Meta, not you. Check metastatus.com (Meta's own business status page) and Downdetector before you change a single setting. If those are lit up red, the problem is not your configuration.
  • Do not reconnect or re-authorize anything. The instinct during an outage is to disconnect and reconnect your account to "fix" it. Resist that. Reauthorizing while Meta's login systems are broken can leave you in a worse state than waiting. Let it recover first.
  • Do not delete or rebuild your flows. Nothing is wrong with them. They will start working again on their own.
  • Wait, then verify. Once status pages go green, send yourself a test DM to confirm your automation is responding before you assume anything is broken.

Ninety percent of the damage businesses do to themselves during an outage comes from frantically "fixing" something that was never broken. The single best move is usually to do nothing and let Meta sort its own house out.

Bottom Line

Meta will go down again. It does every few months, and June 12 will not be the last time. No automation tool, ours included, can deliver a message while Meta's own pipes are offline, and anyone who tells you otherwise is selling something.

What you can actually control is whether your tool loses those messages or just delays them, and whether there is a real person you can reach when you are staring at a wall of error reports wondering if it is your fault. That is the part we care about most. If you ever hit a day like today and want someone to look at your account and tell you straight whether it is Meta or you, message us. A founder will answer.

Frequently Asked Questions

Was there a Meta outage on June 12, 2026?

Yes. Starting around 9:30 a.m. ET on June 12, 2026, Facebook, Instagram, and Messenger had a major global outage, with over 100,000 user reports logged on Downdetector within the first half hour. Meta's business tools, including Ads Manager, Meta Business Suite, the Messenger Platform, the WhatsApp Business Platform, and the Graph API, were also disrupted. Most services recovered within a few hours.

What caused the June 2026 Meta outage?

Meta did not publish an official cause. Based on the symptoms (users being logged out and unable to sign back in with the correct password, with authentication errors rather than network errors), the failure was consistent with a problem in Meta's backend login infrastructure, similar to its March and December 2024 outages.

Did my DM automation break during the Meta outage?

Your automation itself almost certainly did not break. During a Meta outage, the Send and Graph APIs that every messaging tool relies on stop accepting calls, so no tool can deliver Instagram or Facebook DMs until Meta recovers. Your flows and settings stay intact. Once Meta is back, a well-built tool sends its queued replies automatically.

Do I lose leads when Meta goes down?

Not necessarily. Meta retries undelivered inbound webhooks on its side for hours, so messages your customers sent during the outage are re-delivered once the connection recovers. On the outbound side, it depends on your tool: a platform that queues and retries sends will deliver those replies late rather than losing them, while a fire-and-forget setup can drop them.

Should I reconnect my account during a Meta outage?

No. Reconnecting or re-authorizing your account while Meta's login systems are down can leave you in a worse state. Confirm the outage on metastatus.com or Downdetector, leave your settings alone, wait for the status pages to go green, then send a test message to verify everything is working.

Sources