Managing a Digital Crisis: The First Hour
The site is down. The ad account is suspended. A review is spreading. ⚠️ Decisions made in a crisis can undo months of work in a single day — and the worst ones are usually made in the first hour.
The crisis itself is unavoidable; every business meets one eventually. What makes the difference is preparation: when it’s clear in advance who does what, a crisis stays an incident. 🛡️
This guide covers three things: which crises occur, what to do in the first hour, and what to prepare beforehand. 📋
Most crises are really problems of ownership and preparation. Whose name the accounts are in is covered in our contract guide; a crisis is the moment that clause proves its worth. 🔑
The First-Hour Rule 🕐
The most expensive mistake in a crisis is making a broad move in panic. Changing hosting because the site is down, closing social accounts because of a bad review — these don’t solve the problem, they enlarge it.
There’s only one thing to do in the first hour: understand the situation correctly. 🔍
Five Crisis Types and the First Move 📋
BU BÖLÜMÜN ÖZETİ
- When the site goes down
- When the ad account is suspended
- When a negative review arrives
- When traffic drops
Crises fall under five headings. Each has a different first move, and not knowing it costs time.
Printing this table and keeping it somewhere is the most practical preparation there is. 🖨️
| Crisis | First move | Don’t |
|---|---|---|
| Site is down | Check hosting and domain status | Migrate in panic |
| Ad account suspended | Read the notice, appeal | Open a new account |
| Wave of negative reviews | Find the cause, respond individually | Try to delete them |
| Traffic drops suddenly | Check whether it’s real or tracking | Change things immediately |
| Lost access | Start the recovery process | Set up a new account |
When the site goes down
Find the cause first: hosting, domain expiry or a technical fault? 🔌 Domain expiry is the most common cause and the fastest to fix; migrating in panic delays the fix by days.
When the ad account is suspended
The notice usually states the reason. 📧 Work the appeal process; opening a new account compounds the violation and can lead to permanent closure.
When a negative review arrives
Trying to have it removed almost always makes things worse. 💬 The right move is a calm, specific response that offers a resolution — everyone is reading, and the response gets discussed more than the review.
When traffic drops
Check measurement first. 📊 A broken tag looks exactly like a real drop; changes made without knowing the cause can produce a real one. The order is: tags first, technical second, content last.
Four Things to Prepare in Advance 🛡️
BU BÖLÜMÜN ÖZETİ
- Automatic, off-server backup
- A current access list
- A phone chain
- A one-page plan
Preventing crises isn’t always possible, but limiting damage is. Four preparations keep most crises at the level of an incident.
All four take a day, none are expensive. ✅
Automatic, off-server backup
A backup sitting on the same server isn’t a backup. 💾 It should run automatically, live somewhere else, and once a year actually be restored and tested.
A current access list
Which account is in whose name, who can reach it. 🔑 If you’re looking for this list during a crisis, you’re already too late; it belongs in the contract annex from the start.
A phone chain
Who calls whom, in what order. 📞 Who decides if the decision-maker can’t be reached — that question needs answering in advance, not mid-crisis.
A one-page plan
The table above, the access list and the phone numbers. 📄 Keep it printed; when the site is down, a plan that lives on the site is no plan at all.
Handling a Communication Crisis 💬
BU BÖLÜMÜN ÖZETİ
- Fast but not hasty
- Don’t get defensive
- Make the resolution public
- Fix the source
Technical crises get fixed and forgotten; communication crises leave a trace. That’s why they need a different approach.
Four principles. 🎯
Fast but not hasty
Silence reads as an admission; a rushed reply usually enlarges the problem. ⏱️ The ideal is a brief first response, followed by detail once you’ve looked into it.
Don’t get defensive
Even when you’re right, the party that wins the argument loses reputation. 🤝 Everyone is reading, and what they remember isn’t who was right but who behaved well.
Make the resolution public
Complaints happen in public, resolutions in private. 📢 Reverse it: write the resolution where it’s visible; it’s a message to the hundreds of people reading.
Fix the source
If the same complaint recurs, the problem isn’t communication — it’s the service. 🔧 Review management shouldn’t paper over the real issue.
After the Crisis 🔄
BU BÖLÜMÜN ÖZETİ
- Write down what happened
- Make the permanent fix
- Update the plan
- You don’t have to handle it alone
The work doesn’t end when the crisis does. Three things done afterwards make the next one considerably smaller.
An hour’s work. 📝
Write down what happened
Cause, duration, impact and resolution. 📄 Don’t rely on memory; when the same thing happens in six months, this note gives you the fastest fix.
Make the permanent fix
The fix applied during a crisis is usually temporary. 🔧 The permanent one should happen while things are calm, or the same crisis returns.
Update the plan
If what just happened wasn’t in the plan, add it now. 📋 Each crisis makes the plan slightly better for the next one.
You don’t have to handle it alone
Having a technical party reachable during a crisis resolves most problems in minutes rather than hours: Digital Consultancy. To assess how crisis-resilient your current setup is: Digital Audit. 🚀
Frequently Asked Questions 💬
Sık Sorulan Sorular
Establishing scope. Is everyone affected, or just you? 📱 Check from a different device and connection; sometimes the problem is only in your browser. It takes two minutes and prevents half the unnecessary panic.
The decision-maker and whoever handles the technical side. 📞 Don’t make a wide announcement; while nobody knows what’s happening, the information that spreads is usually wrong.
When it started, what’s visible, what the error says. 📸 Take screenshots; once the problem is fixed the evidence is gone and the cause can’t be found.
Multiple changes at once. ❌ Each intervention should happen singly, or you’ll never know which one worked — or which one made things worse.
Understand the situation. Establish scope, inform the decision-maker, take screenshots — and don’t make multiple changes at once.
Find the cause first: hosting, domain expiry or a technical fault. Domain expiry is the most common and fastest to fix; migrating in panic delays everything.
Read the notice and work the appeal process. Opening a new account compounds the violation and can lead to permanent closure.
Trying to almost always makes things worse. The right move is a calm, specific response that offers a resolution.
Check measurement first. A broken tag looks exactly like a real drop; changes made without knowing the cause can produce a real one.
Four things: automatic off-server backup, a current access list, a phone chain and a printed one-page plan.
Not if they sit on the same server. They must live elsewhere and be actually restored and tested once a year.
Four principles: fast but not hasty, don’t get defensive, make the resolution public and fix the source.
Three things: write down what happened, make the permanent fix and update the plan. An hour’s work that shrinks the next crisis.