The worst time to decide how you handle a crisis on social media is during one. When something goes wrong, an outage, a bad press story, a tone-deaf post, an angry customer thread going viral, the clock is loud and the instinct is to either say nothing or say too much too fast. Both make it worse. Crisis communication on social media is mostly preparation: the plans you write on a calm Tuesday so that on a bad day you are executing, not improvising.
This is not about content marketing. It is about protecting trust when the normal calendar stops being appropriate. The teams that come out of a crisis with their reputation intact are almost always the ones who had a plan before they needed it.
First, define what counts as a crisis
Not every complaint is a crisis, and treating them all as one burns out your team and your audience. Sort incidents into tiers so the response matches the reality.
| Tier | What it looks like | Response |
|---|---|---|
| Tier 1 — Noise | A single unhappy customer, an isolated complaint | Normal support reply, log it |
| Tier 2 — Issue | A recurring complaint, a small pile-on, a minor mistake | Acknowledge publicly, pause related promos |
| Tier 3 — Crisis | Outage, safety, data, viral backlash, press attention | Activate the plan: pause, holding statement, escalate |
The point of tiers is speed. The person who first sees the problem should be able to classify it in seconds without waking up leadership for a Tier 1. This is the same discipline behind broader social media governance : clear rules so people can act without asking permission for everything.
Build the plan before you need it
A usable crisis plan fits on one page and answers four questions:
- Who decides? Name the person who can declare a Tier 3 and approve public statements. Name a backup, because crises ignore office hours.
- Who can speak? During a crisis, the number of people allowed to post should shrink, not grow. Lock the queue to a small, named group.
- What do we say first? Pre-write holding statements (below) so the first response is minutes, not hours.
- Where do we coordinate? A single channel where the team aligns before anything goes public, so two people do not post contradictory things.
Keep this somewhere findable at 11pm, not buried in a drive nobody can locate. The reviewers who must sign off in a hurry should already be on your social media approval workflow .
Write holding statements in advance
A holding statement buys you time without committing to facts you do not yet have. It does three things: acknowledges the issue, shows you are on it, and says when you will update. You can draft these now, before any specific crisis, because the structure does not change.
We are aware that some customers are having trouble accessing [service] right now. Our team is investigating and fixing it. We will post an update within 30 minutes. Thank you for your patience.
Notice what it avoids: blame, speculation, “we apologise for any inconvenience” filler, and any promise about cause until you know it. Have variants ready for an outage, a product issue, and a PR situation. The cardinal rule is to never speculate on cause or numbers in public before they are confirmed.
Pause the calendar immediately
The fastest own-goal in a crisis is a cheerful scheduled post going out while customers are furious. The moment you hit Tier 3, pause every scheduled post across every channel. A jokey promo landing mid-outage reads as either oblivious or callous, and screenshots of it outlive the crisis.
This is exactly why a publishing setup needs a one-action pause, not a frantic scramble through six tools to find every queued item. A tidy social media publishing workflow makes this a single decision. Resume the calendar only when the situation is genuinely resolved, and review the first few posts back manually.
Respond in public, resolve in private
The pattern that works: acknowledge publicly so onlookers see you are present, then move the detailed resolution to DMs, email or support. Public threads escalate; private channels de-escalate. Reply to the customer, not the audience, but reply where the audience can see you showing up.
Watch the signals while it unfolds. Set up social listening for content ideas in calmer times so you already have the monitoring in place to track sentiment, spot a second wave, and know when the conversation is actually cooling.
After it is over, do the review
When the crisis ends, the work is not finished. Run a short, blameless review within a few days while memory is fresh:
- How long from first signal to first public response?
- Did the holding statement hold, or did we improvise?
- Did anyone post who should not have been able to?
- What turned the temperature down, and what turned it up?
- What do we change in the plan now?
The output is a sharper one-page plan and, often, a couple of FAQ updates so the next version of this question gets answered before it becomes a thread. Feed those lessons into your content review checklist so the fixes stick.
Utin is being built around website context, which helps here in a quiet way: when your approved statements, FAQs and escalation notes live alongside your content, the team can pull accurate, on-message language fast instead of writing under pressure. You can register interest in the early pilot from the sidebar.