Field Notes · Operations & Automation

Disruption Playbooks: Re-planning a Trip Before the Shuttle Leaves

Most disruptions announce themselves hours in advance. The operators who handle them well have already decided what they'll do.

  • September 15, 2026
  • 8 min read
  • Nova Adventure Field Team
A stagehand cranks a painted backdrop from sunshine to rain while small actors keep performing.
Field noteGood disruption handling is a scene change the audience barely notices.

New Mexico’s summer monsoon brings afternoon thunderstorms. Anyone who has run trips in the high country in July knows the rhythm: clear mornings, clouds building over the peaks after midday, and lightning on exposed ridges by mid-afternoon. It’s about as predictable as disruption gets, and it still catches operators out every season.

What follows is a composite drawn from corporate offsite work near Santa Fe. The times and numbers are typical of what we see. The people are left anonymous.

Thirty-eight people are on a three-day offsite in late July. Day two looks like this:

  • 8:30–12:00 Strategy workshop at the hotel
  • 12:15 Lunch
  • 1:15 PM Two shuttles to a high ridge trailhead, 50 minutes away
  • 2:15–4:45 Guided ridge hike
  • 5:30 Catered dinner at an open-air overlook
  • 8:30 Shuttles back

Day three has a free morning, a closing lunch at noon and departures from 2:00 PM.

The timeline

6:15 PM, the day before. The forecast gives a 40% chance of thunderstorms over high terrain after 1:00 PM tomorrow. The playbook for afternoon storms on exposed terrain moves to watch. An exposure check flags two items: the ridge hike and the open-air dinner. Travelers hear nothing. Two suppliers get pre-alerts. One is the caterer, whose contract includes a covered courtyard alternate on six hours’ notice. The other is the shuttle company, which needs three hours’ notice for changes.

6:00 AM. The updated forecast is 70%, with cells building from about 1:40 PM. That crosses the act threshold. The routing engine produces four options, each with its cost, the exposure that remains, and the time the option stops being available:

Option What changes Added cost Exposure left Available until
A. Go early Hike 11:00–1:00, workshop cut short Shuttle change fee Ends 40 minutes before forecast onset, no margin 7:10 AM
B. Swap days Hike moves to tomorrow’s free morning, 7:30–10:00; this afternoon becomes a clay studio session in two groups; dinner moves to the covered courtyard Courtyard and studio fees None 9:30 AM
C. Replace Hike cancelled; studio sessions and courtyard dinner Studio fee, forfeited guide fee None 9:30 AM
D. Hold Keep the plan; the guide decides at the trailhead None yet High Always

6:20 AM. The operations lead reads the options on her phone. B is clearly best. It leaves no exposure at a modest cost, and the only thing lost is a free morning that half the group would have spent on email. She chooses it at 6:24. At 6:40 she calls the client’s event lead, who agrees within two minutes because the options arrive already costed.

6:45 AM. Change requests go out. The playbook drafted them and the operations lead approved them in one step. The caterer moves dinner to the courtyard. The shuttle company cancels the 1:15 ridge run and adds a 6:50 AM pickup tomorrow. The studio confirms sessions at 2:00 for 24 people and 3:45 for 14. The guide moves to tomorrow at 7:30. All four confirm by 8:05.

8:15 AM. One message goes to all 38 attendees before the workshop starts. The client’s event lead gets it ten minutes earlier.

1:15 PM. The shuttles don’t leave for the ridge.

1:50 PM. Thunder rolls over the high country.

7:30 AM the next day. The group is on the ridge under a clear sky.

None of that took heroics. It took decisions made in advance: what each forecast number means, which alternates are pre-contracted, who decides, and by when. That’s what a playbook is.

Anatomy of a playbook

Every playbook we write has the same seven parts, and each fits on one page:

  1. Trigger. A named data source, a watch threshold that starts internal preparation, and an act threshold that starts the re-plan.
  2. Scope. A rule for finding what’s exposed. For example: any activity on open terrain between noon and 6:00 PM, and any open-air venue.
  3. Pre-approved options. Alternatives generated against live constraints and pre-agreed supplier alternates, each with its cost and expiry.
  4. Decision owner and deadline. A named role, with a backup if there’s no response within 15 minutes.
  5. Messages. Templates for travelers, the client, guides and suppliers.
  6. Supplier actions. Change requests that carry the contract clause they rely on.
  7. Closeout. What gets logged, and how it adjusts next season’s thresholds.

Triggers and thresholds

These are starting points, not standards. Tune them each season using your closeout logs.

Disruption Signal Watch Act Typical first move
Afternoon thunderstorms Hourly forecast for the activity’s location and window 30% or more inside the window 60% or more, or lightning nearby Move exposed activities to the morning, or swap days
Wildfire smoke Air-quality forecast AQI above 100 AQI above 150 during planned exertion Lower the exertion, move indoors, or change areas
Road or trail closure Agency notices, supplier messages Closure announced for the date Closure confirmed Re-route and brief drivers
Arrival flight delay Flight status Delay over 45 minutes Delay over 2 hours, or a cancellation Re-time the transfer and protect the first-night dinner
Supplier failure Missing confirmation Unconfirmed 48 hours out Unconfirmed 24 hours out Bring in the backup supplier
Extreme heat Forecast high 95°F or higher 100°F or higher Move exertion before 10:00 AM

The last responsible moment

Every option has an expiry, set by the notice its suppliers need. Your decision deadline is the earliest expiry among the parts of the option you’d most like to keep, minus a buffer. For option B:

Part of option B Notice needed Latest change
Courtyard dinner Six hours before the 3:30 PM setup 9:30 AM
Clay studio sessions Tentative block released by the studio 10:00 AM
Cancel the 1:15 PM shuttle run Three hours 10:15 AM
Guide moves to tomorrow Same-day agreement Flexible

With a 15-minute buffer, the deadline was 9:15 AM. So why decide at 6:24?

Because the last responsible moment is a limit, not a goal. Waiting only pays if the information is likely to improve. At 6:00 AM, with a 70% forecast and a clear onset time, the 9:00 update was unlikely to change the choice, and every hour of waiting cut into the attendees’ notice. Decide when the information stops improving or the deadline arrives, whichever comes first.

Option A teaches a second lesson. It was gone by 7:10 AM. Some options expire early, so if you want early-start alternatives, the watch phase has to begin the evening before.

Pre-negotiate the alternates

Option B worked because the alternates were agreed in April, not improvised in July. Useful clauses for the monsoon season include:

  • Caterer: a covered alternate at a fixed fee, on six hours’ notice.
  • Shuttle company: free re-timing within 24 hours, given three hours’ notice.
  • Indoor venue or studio: a tentative hold on monsoon-season afternoons, released at no cost by 10:00 AM.
  • Guide: a weather clause that pays a reduced fee for a moved or cancelled hike rather than nothing. That keeps good guides loyal.

Store each clause with the supplier record, so that a change request cites the clause it relies on. That is what lets automated supplier requests go out in minutes instead of turning into phone negotiations.

One message, sent once

During the watch phase, travelers hear nothing. When the plan changes, they hear it once: the new plan first, the reason in a sentence, and what they need to do.

Change to this afternoon. Your new plan is below.

The ridge hike moves to tomorrow at 7:30 AM (shuttle from the lobby at 6:50). Thunderstorms are forecast over the high trails from about 1:40 PM today, and we don’t take groups onto exposed terrain in that window.

This afternoon: clay studio sessions at 2:00 PM (Group A) or 3:45 PM (Group B). Your group is printed on your badge. Dinner is at 6:00 PM in the covered courtyard.

Nothing to do unless you’d rather skip tomorrow’s hike. Reply “skip” and we’ll arrange a late breakfast instead.

Then come the follow-up questions: Can I switch to Group B? Is the courtyard far? This is where an in-trip concierge earns its keep. It answers from the new plan and hands anything unusual to a person. Three attendees replied “skip,” and each got a late breakfast without anyone picking up a phone.

What to automate and what to keep human

Automate Keep human
Watching forecasts and feeds against every active itinerary Choosing the option
Finding exposed activities and the travelers affected Safety calls on the ground; the guide can always turn back
Generating and costing options with a routing engine The conversation with the client
Drafting messages and supplier requests Exceptions: medical needs, anyone upset
Tracking confirmations and chasing silence Changing the thresholds

The left column is where the minutes go. The right column is where judgment goes. Keeping them apart is what makes re-planning, rather than rebooking, realistic at 6:00 in the morning. That division of labor is the core of Nova’s Disruption Response Automation work. It matters most for corporate travel and events, where one change affects dozens of people and a client is watching. The offsite planning case study shows the same pattern across four live events.

Build your first three playbooks

  1. Count last season’s disruptions. Pull the incident log, or rebuild it from email, and tally them by type.
  2. Pick the three most frequent. For most operators in the Mountain West, afternoon storms will be one of them.
  3. Write the seven parts on one page each. Keep the thresholds simple enough to remember.
  4. Call your five most important suppliers and negotiate alternate clauses before the season starts.
  5. Write the traveler messages now, while nobody is under pressure.
  6. Run a tabletop drill. Take one real day from last season, replay its forecasts hour by hour, and time each step.
  7. Close out every live use with 15 minutes of notes on what fired, what you chose and what you would change.

The aim isn’t to stop the weather getting a vote. It’s to make sure that when it does, the shuttle sits in the lot and your group already knows where it’s going instead.

#disruption management#re-planning#tour operations#monsoon weather#group travel

Questions

Questions readers ask

How far in advance should a disruption playbook trigger?

Use two thresholds. A watch threshold, typically 18 to 24 hours ahead for weather, starts internal preparation and supplier pre-alerts. An act threshold, tied to the deadlines of your options, triggers the re-plan. The right lead times come from your suppliers' notice periods, not from the forecast.

Should travelers be told when a disruption is only possible?

Usually not. Keep the watch phase internal and tell travelers once, when the plan has actually changed, with the new plan first. The exception is anything travelers must prepare for themselves, such as packing rain gear or setting an earlier alarm.

Keep reading

Related field notes

Turn this note into a plan for your operation

Bring the problem this article describes to a 45-minute discovery call. We will tell you honestly whether AI is the right tool for it yet.