The T-minus timeline (T-48h through T-0)

Hurricane preparation for a vacation rental operator isn't a single decision — it's a 48-hour cascade of decisions, any one of which can cost thousands if it's late. The operators who weather storms well aren't the ones with the smartest people. They're the ones whose systems are pre-decided, so when the cone drops on the market the sequence runs itself and the humans spend their time on the calls that need human warmth.

Here's what that 48-hour cascade looks like when the Ops agent, the Guests agent, the Revenue agent, the Trust agent, and the Align agent are all running coordinated.

T-48 hr — Storm SOP activated. The moment the cone updates and your properties fall inside the forecast path, Schedule.guard fires the storm SOP. That's not a memo. That's a set of parallel actions: 31 units flagged as in-cone, vendor board-up calls dispatched to the hurricane vendor network, interior inventory staging tasks created in Breezeway for every affected property, and the storm-mode task board posted for the market lead. No meeting required. The SOP existed in your Ops agent's playbook library, pre-approved, and it executed the moment the trigger condition hit.

T-47 hr — Rebook and refund logic applied. Recovery.planner walks the reservation list and applies the per-channel policy to every in-window booking. Airbnb reservations get the Extenuating Circumstances treatment (full refund, no charge to the owner). Vrbo reservations get the partial-refund policy. Direct-booked guests get the full-refund path consistent with your direct terms. Every guest gets a pre-drafted notification ready for review. The guest communications lead reviews, edits for tone, and ships the batch — 31 messages out in an hour, not a week.

T-46 hr — Dynamic pricing frozen. Mix.optimizer locks the pricing engine into manual-override mode for every unit in the affected market. This is the T-minus move that operators chronically miss: storm-driven demand spikes (evacuation traffic, media crews, insurance adjusters) can push PriceLabs to exploit-level rates on units that are about to lose power or roof. Freezing prevents the reputation damage of algorithmically-set storm-peak pricing and prevents the inventory-misalignment of selling nights on units about to be uninhabitable. The rates stay where they were. No manager has to remember to freeze them. The freeze is the default state once the SOP activates.

T-40 hr — Insurance reserves pre-allocated. The Trust agent's Trust.ledger opens 31 per-owner insurance reserves. Claim templates are pre-populated with property addresses, owner contact, insurance carriers (where on file), and photo-inventory references. If the storm hits as forecast, the first insurance call can happen within 24 hours of landfall, not three days later. The ledger move is quiet, but it's the one that saves weeks on the back end.

T-36 hr — Owner portfolio brief sent. Align agent's Forecast.updater produces the per-owner impact summary — which of your owner's properties are in the cone, what the SOP is doing for each, the insurance posture, the booking impact, the expected revenue loss. Each owner gets their portfolio brief automatically, cover note drafted for review by the owner relations lead. The 31 owner calls still happen, but they happen after the owner has the summary in their inbox — so the call is a conversation about the plan, not a panicked "what's happening to my property."

T-24 hr — Final staging and seal-off. Vendors confirm board-up completion. Interior staging photos captured and tied to each property's task record. Water heaters shut off where applicable. Generators pre-staged at priority properties. The task board shows every unit's readiness status, green or flagged.

T-12 hr to T-0 — Monitor and hold. By now the humans are out of the market or in their storm-safe posture. The agents keep watch: sentiment drops in any remaining inbound messages, owner updates ready for the post-landfall batch, insurance claim drafts on deck for the moment the damage assessment comes back.

The whole 48-hour cascade runs off one SOP trigger, coordinated across five agents. The operator isn't running the storm response by manual push. They're reviewing the decisions the SOP is making and correcting the exceptions.

Storm SOP: vendor board-up, inventory staging

Two pieces of the SOP are worth breaking out because they're where the manual version burns the most hours.

Vendor board-up. Most operators have a preferred board-up vendor or two per market. The manual flow is: someone on the ops team pulls the list of affected units, calls the vendor, negotiates availability, dispatches addresses, and tracks completion over text messages and phone calls. With 31 units, that's a full day of coordination for one person. The Ops agent replaces that coordination with a pre-built task queue: every in-cone property generates a board-up task with the property address, the unit's board-up profile (window count, door count, exposure notes), and the vendor auto-assigned based on the market. The vendor sees the queue, takes the jobs, posts completion photos. The ops lead sees completion counts rolling up live instead of fielding 31 status texts.

Inventory staging. Hurricane-exposed units need interior prep: patio furniture brought in, loose items secured, refrigerators adjusted if power's likely to go out, personal items cleared from ground floors, vehicle moves where owner vehicles are on property. Each of those is a task template in Breezeway. The SOP fires the full bundle the moment the cone enters the market. The house-prep vendor works through the list in priority order. Every completed task posts a photo, which gets saved against the property — providing the baseline documentation for any post-storm insurance claim.

Per-channel rebook/refund policy (Airbnb XC, Vrbo partial, direct full)

The refund mechanics that the manual teams get wrong are almost never about malice. They're about time pressure under bad UI. Each channel has its own policy, its own refund mechanics, and its own window. Airbnb's Extenuating Circumstances path is the cleanest — once the storm qualifies as an EC event, the full refund applies without the owner absorbing it, but you have to trigger the EC flow correctly. Vrbo typically requires a partial-refund handling negotiated against the host cancellation policy. Direct bookings are governed by your own terms of service, which usually allow a full refund for a named storm.

Run that logic across 31 reservations on 31 different channel-policy combinations under 48-hour time pressure, and the manual team makes mistakes. Mistakes in this context are owner-visible: one reservation refunded from the owner's reserve when it should have been the channel's responsibility, or vice versa. Each mistake is a $400-$3,000 argument with an owner.

The Recovery.planner applies the policy per reservation, pre-computes the refund split (channel side vs. host side), drafts the guest message, and queues the batch for one-click send. The operator reviews the split-list as a table, confirms, and ships. Zero policy misapplications because the rules are in code, not in a teammate's head at 2 AM.

Dynamic pricing freeze — the peak-during-storm exploit

This deserves its own paragraph because it's the thing most operators learn the hard way. Dynamic pricing engines don't have a "storm" concept. They see demand spikes. Demand spikes drive prices up. When a hurricane is 48 hours out and evacuation traffic starts hitting the booking funnel, your rates can spike 3-5x on units that are about to lose roof.

The bad versions of this show up on local news. "Property manager price-gouging evacuees" becomes a brand story, and every owner in that market gets the phone call. Even if your rates weren't gouging by absolute dollars, a storm-time spike is a perception problem you don't want.

The Mix.optimizer storm-mode freeze is boring on purpose. The second the storm SOP activates, prices lock where they are. Any price changes require human override with a logged reason. The audit trail is clean. The local news story doesn't happen because the rates didn't move.

Owner portfolio brief — per-owner impact before the phone trees

The owner-facing piece of the storm response is the piece that separates professional operators from amateur ones. When a hurricane is 36 hours out, owners are anxious, and the operators who make them less anxious win the relationship for years.

The anti-pattern is the mass email: "All owners — we're activating our storm protocols, please stand by for updates." That email is worse than no email, because it tells the owner nothing about their property specifically.

The pattern that works is the per-owner brief. Owner Smith gets a personalized summary: "You have three units in the cone — Smoky Ridge #114, Smoky Ridge #208, and Ridge Top #17. Board-up is dispatched to all three with completion expected by 9 AM tomorrow. Insurance reserves are pre-allocated at $5,000 per unit. Reservations in the 72-hour window have been proactively cancelled per channel policy — $4,100 of booking revenue refunded, $2,400 of that from channel-side compensation. Post-storm assessment will begin within 24 hours of landfall; I'll call you by Tuesday afternoon at the latest." Signed by the owner relations lead.

The Forecast.updater assembles the per-owner brief from live data. The owner relations lead reviews each one, adds the personal note, and sends. The 31 calls happen the next morning, and every call starts with "I saw your brief, thank you." The calls take 15 minutes instead of 45 because the owner already knows the status.

What this looks like when the storm actually hits

Landfall is now a known condition, not an emergency. The team is out of the market or in its safe posture. The board-up is done. The rates are frozen. The reservations are handled. The insurance reserves are pre-allocated. The owners know what's happening. The ops team's job during the storm is to monitor and respond to exceptions, not to run the storm response on fumes.

Post-storm, the damage assessment flows into the insurance claim templates, which flow into the per-owner ledgers. The three-way match resumes the day after the power comes back. The trust variance stays under a dollar because nothing about the storm created reconciliation gaps — every action wrote to the ledger as it happened.

Ready to see the storm playbook against your portfolio? Request access →