Your event platform goes down on event day, the doors open in twenty minutes, and the registration screen is showing a spinning wheel of doom. Here is the short version of what to do: switch to offline check-in if your tool supports it, get a human on the phone at your vendor immediately, keep selling and admitting people manually, and communicate calmly with your team. The long version, with a checklist you can actually run while your heart rate is doing something unfortunate, is below.
This is the nightmare every organiser quietly rehearses in the shower. It is also more common than the glossy sales demos suggest. The difference between a five-minute wobble and a ruined morning is almost entirely about preparation and about whether there is a real person at your platform who will pick up the phone when it counts.
First 60 seconds: triage, do not panic-refresh
Resist the urge to mash F5 like it owes you money. Work out fast what has actually broken, because "the platform is down" usually means one specific thing has failed, not all of them.
Is it your internet or the platform? Try loading any other website on the venue Wi-Fi. If nothing loads, the problem is the venue, not your software, and your fix is different (more on that below). Is it check-in that has failed, or payments, or the whole dashboard? Pin down the symptom before you pick a remedy. And check the obvious: is it just one device or laptop misbehaving, in which case grab a different one and carry on.
If the Wi-Fi died: this is why offline mode exists
Venue Wi-Fi failing under load is the single most predictable disaster in events, which is exactly why serious check-in tools cache the attendee list locally on each device. Platforms such as Eventleaf, Micepad, InEvent and Eventify store the full attendee list on the device so scanning, name search and badge printing keep running with no connection, then sync automatically when the network returns.
If your tool has an offline mode, the move is simple: switch every check-in device to offline, keep scanning, and let it reconcile later. If your tool does not cache locally and depends on a live connection for every scan, you have just learned something important about your platform choice, and your fallback is the manual one below.
The reason offline-first matters so much is that check-in is the one moment where every attendee converges at once, so a thirty-second stall per person becomes a thirty-minute queue in a hurry. The platforms that survive are the ones designed so that the phone in a steward's hand already holds everything it needs to admit someone, with or without a signal. If you want to see how that scan-and-search approach holds up when a thousand people arrive in the same window, we walked through it in this guide to checking in 1,000 attendees on phones. The principle is the same whether you are admitting fifty people or five thousand: resilience is a design decision made long before the doors open.
"Works perfectly" and "works perfectly when the Wi-Fi is good" are two completely different promises. Only one of them helps you at 8am in a concrete basement.
If the platform itself is down: get a human, fast
This is the moment that separates platforms. When something breaks live, you do not want a chatbot, a ticket queue, or a "we will respond within 48 hours" auto-reply. You want a person.
It is worth being honest that the big enterprise tools are not immune here. Reviewers of Cvent have reported features not working on "game day" and struggling to reach their assigned support contact when it mattered, with a support structure spread across multiple people that made clear answers hard to get. Cvent does offer an onsite phone-callback option through its community profile, so the channel exists, but "the channel exists" and "a human answered in ninety seconds" are not the same experience when there is a queue forming behind you.
The lesson is not "Cvent bad". It is that you should know, before event day, exactly how you reach a live human at your platform during a live event. Is there a direct line? A named contact? An emergency channel? If the honest answer is "I would email support and hope", that is a gap to close now, while it is cheap to close.
The manual fallback that always works
Technology fails. Clipboards do not. Every event should have an analogue plan B that needs no power and no signal:
Keep an exported attendee list (a simple spreadsheet or printout) on a laptop or on paper, refreshed the night before. If check-in dies, tick people off the list by name and admit them. Note walk-ins and on-the-door sales on paper to enter later. If card payments fail and you sell at the door, have a backup card reader or a clearly communicated alternative ready. None of this is glamorous, but a moving queue beats a perfect database every single time. You reconcile the data afterwards.
The deeper protection is structural. If your money runs through your own payment account rather than being held by the platform, a platform wobble does not freeze your funds, and your existing sales are not hostage to someone else's server. That is a design choice worth checking when you pick a tool, not when one is on fire.
Print the analogue plan and keep a paper copy in the registration kit, not just a file on a laptop that might also be having a bad morning. A laminated single page with the attendee list, the vendor emergency contact and the three steps your team should take buys you a remarkable amount of composure. The crews who sail through outages are rarely the ones with the most expensive software. They are the ones who decided, in advance and in writing, what to do when the expensive software stops cooperating.
Keep your team and your attendees calm
Half of crisis management is communication, and the audience is watching how you behave. The instant something breaks, tell your team in one short message what is happening and who is doing what, so you do not have six people all phoning the same support line while nobody mans the door. Nominate a single point of contact for the vendor before the event so there is no confusion about who makes the call.
For attendees, a calm holding line works wonders: "We are checking everyone in manually for a few minutes, thanks for your patience, the bar is open." People forgive a short, well-handled delay almost instantly. What they remember is panic, silence, or a steward loudly blaming "the system". Keep the tone light, keep the queue moving, and most attendees will never know how close the morning came to falling over. Brief your front-of-house team on that script in advance, because the worst time to write it is mid-meltdown.
A pre-event resilience checklist
You cannot prevent every outage, but you can make one a non-event. Before your next event, confirm the following: your check-in app has a tested offline mode and you have actually tried it with the Wi-Fi off; you have an exported attendee list living somewhere that does not need the internet; you know the exact, fastest way to reach a live human at your platform during the event; you have a backup device and a backup payment method; and your team knows who does what the moment something breaks. Run that list once and event day gets a lot less terrifying.
What "good" looks like in a platform
When you are choosing or reviewing an event platform, weight these resilience criteria as heavily as the shiny features: local offline caching for check-in, payments that land in your own account rather than the platform's, reliable automated comms so attendees are not stranded, and genuine human support you can reach during a live event. A tool that nails those will save you on the one morning that actually matters. eventcloud was built around exactly that combination, but the point stands whoever you choose: judge your platform on how it behaves when things go wrong, because eventually they will.
Outages are not really a technology question. They are a "who has your back at 8am" question. Pick a platform that answers it well, run the checklist, and the spinning wheel of doom becomes a brief, forgettable hiccup rather than the story everyone tells about your event. See how eventcloud handles check-in, payments and support if you want resilience built in rather than bolted on.
