Is ManyChat Down? What 155 Outages Since 2018 Actually Show

If you searched “is ManyChat down,” you're probably mid-campaign, staring at an inbox that stopped moving, wondering whether the problem is your account, your setup, or the tool. So we did the boring, useful thing: we read every incident ManyChat has published on its public status page and sorted them by what actually caused each one.

The catalogue runs to about 155 incidents between November 2018 and July 2026. That number sounds alarming until you look at what's underneath it — because the single most important finding isn't about ManyChat's engineering at all. It's about a dependency that every tool in this category shares, including ours. Let's get into it honestly.

The breakdown: what actually caused 155 incidents

When you tag each incident by root cause, the distribution is lopsided in a way that reframes the whole question. The overwhelming majority trace back to the platforms ManyChat sits on — Meta's Facebook, Instagram and WhatsApp APIs — going down, throttling, or shipping bugs of their own.

Read the top row twice. Roughly 74% of everything that ever went wrong — about 115 of the 155 incidents — originated on Meta's side of the fence: the Instagram Messaging API, the Facebook Graph API, or the WhatsApp Business Platform having an outage, elevated error rates, or a breaking change. That is not ManyChat's code failing. That's the ground moving under it.

The genuinely ManyChat-owned buckets — its own infrastructure, feature bugs, data and billing — add up to roughly a quarter of the log combined. For a platform that's processed messages for, by its own published claim, over a million businesses across 190 countries since 2015, that's a fairly ordinary operational record. The story here was never “ManyChat can't keep the lights on.”

The pattern that should shape your planning

Within that 74%, one sub-pattern matters more than any other for a working merchant: every four to six months, on average, a major Meta-wide outage takes down all messaging channels at once. Not a partial slowdown — a simultaneous failure of Instagram, Messenger and WhatsApp because the shared plumbing behind them broke. These are the days the whole internet notices, not just you.

A sample of the documented big ones, straight from the log:

Around those headline events sits a much denser layer of smaller ones: partial degradations and “elevated error rates” that come and go over hours. There are dozens across 2019–2025 — clusters in October 2023 and October 2024 stand out — where sends slowed or a fraction of messages failed before Meta quietly recovered. Most never trend on X. All of them can land in the middle of your launch.

The reliability question for DM automation was never “is this vendor good at engineering?” It's “what happens to my sales the next time Meta has a bad afternoon?”

What this means when it's your launch day

Put yourself in the specific moment this matters. You've spent on a creator collab, the reel is climbing, comments are filling with “price?” and “link?”, and the automation that's supposed to catch every one of those hand-raises… goes quiet. Not because you misconfigured a keyword. Because Meta's Instagram Messaging API is returning errors for everybody, and no DM tool on earth can send through a door Meta has bolted shut.

The uncomfortable truth the data forces is that your exposure to this is roughly the same whatever DM tool you run, because they all send through the same Meta APIs. Switching vendors to escape Meta outages is like switching airlines to escape the weather. What actually changes your outcome is how you plan around a dependency you can't remove:

• Assume a Meta-wide outage will land on one of your big days each year — the base rate says it will, eventually. Have a manual fallback you can run for a few hours: pinned story, link in bio, a human watching the inbox.
• Don't schedule your single most important drop into a known-fragile window if you can avoid it, and keep an eye on Meta's own platform status, not just your tool's.
• Judge a DM tool on what it does during the 99% of the time Meta is up — grounded answers, cart recovery, speed — not on a reliability edge none of them can actually offer against Meta itself.
• When messages fail, you want a tool that queues and retries cleanly once the API recovers, rather than silently dropping the conversation.

The part where we implicate ourselves

Here's where an honest teardown has to turn the lens around. It would be easy to end this with “…so switch to HyperDM.” That would also be nonsense, and you'd see through it, because HyperDM runs on the exact same Meta APIs ManyChat does. We connect to Instagram and WhatsApp through Meta's official Graph and Business Platform endpoints — that's the compliant way to do it, and it's the only way we'd want to. Which means when Meta has one of its twice-a-year bad days, our sends are degraded too.

We cannot fully insulate you from a Meta-wide outage. Nobody selling you Instagram or WhatsApp automation can, and anyone who claims otherwise is either misunderstanding their own stack or lying to you. If a Meta blackout is happening, the honest status update from any of us is the same three words: it's Meta's side.

What you can actually control

Strip out the 74% that belongs to Meta and you're left with the part any tool is truly accountable for: its own infrastructure, its answer quality, its handling of the messages that do get through. That's the right place to spend your scrutiny. On the reliability you can't buy your way out of, plan a fallback. On the value you can control — does the bot answer “does this run small?” correctly, does it recover the cart, does it bill you fairly — pick the tool that's actually better.

That's the case we'd rather make. See how HyperDM answers when the APIs are up on HyperDM for Instagram and WhatsApp, and if you want to put a rough number on what a Meta-wide outage costs your specific store, we built a Meta outage impact calculator for exactly that. It won't prevent the outage — but it'll tell you how much of a fallback plan is worth building.

And if the reliability story is what nudged you to reconsider your stack, the switch page lays out the rest of the comparison honestly — pricing, answers, and all.