The setup
Crisis playbooks are written in calm rooms and executed in loud ones. The playbook says “respond quickly and transparently.” It does not say whether quickly means a complete statement at noon or an acknowledgment at noon and the substance tomorrow — and that timing choice, made once under pressure, shapes the entire arc of coverage.
This scenario lets you make the choice twice and keep the better outcome. The setup is a real journalistic pattern: an accurate story with one exculpatory fact missing. The audiences are the ones that matter in the first two days — affected users discovering they’re affected, advocates for whom this is the week’s case study, media deciding whether there’s a second story, and regulators who say nothing and bookmark everything. The horizon is 48 hours because that is the window in which the frame hardens.
Run both response plans as separate simulations against identical starting conditions. What the comparison exposes is the trade every comms lead knows and rarely gets to test: the fast partial statement buys silence at the cost of a “still no answers” follow-up, while the slower full statement risks twelve unanswered hours in exchange for closing the file. Watching each trade play out against the same simulated newsroom is the closest thing to a fire drill this situation allows.
The prompt
Copy this into MiroFish as your scenario question, swapping the specifics for your own:
A journalist publishes at 9am: our fitness app shared workout location data with an ad partner in a way our privacy policy arguably did not cover. The story is accurate but omits that the sharing stopped four months ago. Simulate 48 hours across: affected users, privacy advocates, tech media following the scoop, our own social accounts, and regulators watching quietly. Compare two response plans — a full statement with timeline at noon, versus a short acknowledgment at noon and the full statement next morning. Tell me where each plan puts us at hour 48, and which follow-up story each one invites.
Seed files that help
- The two draft statements, word for word — in a crisis simulation the statement is the intervention, so both versions must be real.
- A factual timeline of what happened and when it stopped, so the simulation knows the truth even though most agents start without it.
- Two or three examples of how similar incidents at other companies played out, which calibrates the media agents' instincts for follow-ups.
What to look for in the report
- Escalation gradient: how fast the story climbs from tech press to general press in each plan — the steepest climb usually follows the vaguest statement.
- The omitted-fact dynamic: when and how 'it stopped four months ago' surfaces, and whether it reads as exoneration or as 'they knew and said nothing.'
- Anger half-life: in which plan user anger peaks earlier and decays faster — speed of peak matters less than what's left at hour 48.
- The invited follow-up: each response strategy makes some second-day story more likely; the report names the one you're choosing.