Crisis or Noise: Auditing a Live Mentions Dashboard

Marketing Academy · Field Work●Social Media Marketing

All projects·From: Social Media Crisis Management & Real-Time Response

MiniAudit· 20 minutes

Yelp

Objective: Given a snapshot of a real-time mentions dashboard, apply the velocity/spread/amplification framework to decide whether a spike qualifies for the full crisis protocol or just a monitoring note.

You're the social listening analyst at Yelp, the local business reviews and discovery platform, watching mentions spike on a Saturday afternoon with nobody senior reachable for 40 minutes.

Score three separate mention spikes against the velocity/spread/amplification framework and decide which one, if any, triggers the full crisis protocol.

Before you start

What you'll need

Free path (everything below is enough to finish)

Google Alerts

FreeBaseline mention monitoring outside business hours when a paid dashboard seat isn't staffed

Free and requires no login to set up a keyword watch

Paid upgrades (optional, faster/deeper)

Brandwatch(optional)

PaidReal-time mentions dashboard with volume, platform, and account-reach breakdown

The velocity/spread/amplification signals in this exercise require a real-time cross-platform view a free tool alone can't match

The process

1 step

Step 01 of 01

Velocity, spread, and amplification together define a real crisis

One signal alone rarely justifies a crisis protocol; two of three signals trigger a manager review, all three trigger the full response.

Three mention spikes are on your dashboard right now. Which one, if any, has all three signals?

Brandwatch— A mentions dashboard export listing volume-over-time, platform breakdown, and top accounts driving each spike.

Procedure

For each of the 3 spikes, check volume trend over the last 2 hours (Velocity)
Check whether mentions are confined to one platform or appearing on 2+ (Spread)
Check whether any account over 50k followers has shared it (Amplification)
Score each spike 0-3 and apply the 2-of-3 rule

Sample output

YELP MENTIONS DASHBOARD, Saturday 2pm snapshot

Spike 1, "slow refund" complaint thread: Velocity flat (11 mentions/hr steady) | Spread: X only | Amplification: none -> Score 0/3, log and monitor
Spike 2, review-bombing on a single listing: Velocity climbing (8 to 95 mentions/hr) | Spread: X + Reddit | Amplification: a 40k-follower local news account shared it -> Score 2/3, manager review
Spike 3, data-privacy claim: Velocity climbing (12 to 340 mentions/hr) | Spread: X, TikTok, Reddit | Amplification: a verified tech journalist (210k followers) quote-tweeted it -> Score 3/3, full protocol

Healthy

Only Spike 3 triggers the full protocol, the score matches the evidence, not gut instinct about which one 'feels' worse.

Unhealthy

Escalating Spike 1 because the language is aggressive, while missing that Spike 3's velocity is climbing 28x faster.

What this means

Score 2/3 spikes still need a human manager decision, they don't auto-escalate and they don't get ignored either.

So what do I do about it?

Symptom	Action	Effort
A loud but flat-velocity thread is being treated as urgent	Check the 2-hour trend line before reacting to tone alone	5 min

YouYou can do this yourself, no engineering access required.

Final deliverable

A scored table of all 3 spikes against the velocity/spread/amplification framework, with an escalation decision and one-line justification for each.

See a reference example

Sample output

Care.com weekend mentions triage

Spike A (background-check turnaround complaint): 1/3 signals, flat velocity, single platform, no amplification -> log and monitor.
Spike B (a safety-incident allegation shared by a parenting influencer with 85k followers): 3/3 signals -> full protocol, manager paged immediately.

Success criteria

You're done when you can:

All 3 spikes scored using all 3 signals, not just the loudest one
Correctly applies the 2-of-3 escalation rule
Justification cites specific numbers from the dashboard, not vague language