You’ll Be the Last to Know
By Rachel Godbout, COO — Advanced Outdoor…
Two summers ago, on a Friday afternoon, a regional manager called me about a property in the Midwest.
The forecast for the weekend had just turned. A system that had been tracking south all week was now sliding directly over the property — three days of heavy rain, low fifties, severe storm risk Saturday night. Reservations for the weekend were at 87% occupancy. The phones were starting to ring.
His question was the one every operator has had a hundred times: What do we do about cancellations?
The honest answer in that moment was the answer most of this industry has used for decades: we’ll handle it case by case, try to be fair, and hope it doesn’t tank the reviews. We’d move some guests to future stays. We’d hold the line on others. The front desk would take the brunt of it. By Tuesday, we’d have lost maybe 30% of the weekend’s revenue, generated a handful of unhappy reviews, burned out a GM, and learned nothing transferable.
That conversation is what made us decide we were done doing it that way.
Two years later, every property in the AOM / RJourney portfolio runs on a proprietary weather-integrated decision system that connects forward-looking forecast data directly to pricing, policy, and operational protocols. We didn’t license it. We didn’t buy it. We built it — because we couldn’t find anyone in the outdoor hospitality space who had built one.
This is what we learned.
Outdoor hospitality is the only major sector of the travel industry where weather is simultaneously the most important variable and the least systematically managed one.
Hotels don’t care if it rains. Cruise lines hedge with onboard programming. Airlines have automated rebooking systems triggered by FAA data. Even ski resorts — which depend on weather more than anyone — have decades of operational protocols built around snowfall data.
Outdoor hospitality, despite living and dying by the weekend forecast, has historically treated weather as a series of one-off conversations. A GM checks the radar Friday morning. A regional manager fields a call from a property in distress. A guest argues with a front-desk clerk about a refund policy that doesn’t account for the actual weather they’re getting. The system, to the extent there is one, runs on judgment, mood, and how busy the front desk is on a given afternoon.
That’s not a system. That’s a coping strategy.
The first real challenge we ran into wasn’t technical. It was definitional.
What is “bad” weather for a property?
The intuitive answer is “rain.” But rain at 78 degrees is a feature for some properties — the kids splash in puddles, the bathhouse fills up, the dog park empties out for an hour, and the property functions normally. Rain at 51 degrees, sustained over 48 hours, with high winds, is something else entirely. It empties the pool, drives families into cabins, breaks down marginal RV setups, and produces the exact guest-experience conditions that generate negative reviews.
The same is true on the upside. “Good” weather isn’t just sunshine. Sustained sunshine over 95 degrees with high humidity creates safety risk, drains pool capacity, and generates as many cancellations as a cold snap. Perfect weather has a specific definition, and it varies by region, season, property type, and guest segment.
We spent the first six months of this project not building software. We were defining, property by property and segment by segment, what “good” and “bad” actually meant in terms of guest behavior, cancellation patterns, and operational stress. The output was a set of property-specific weather quality definitions — not a single global threshold.
That definitional work is what most operators skip. It’s also where most of the value lives.
Once we had the definitions, the system itself follows a relatively simple logic.
The weather data layer pulls forward-looking forecasts at a property-specific resolution, updated continuously, with multiple model agreement and confidence scoring. The interpretation layer maps that forecast against our property-specific quality definitions. The decision layer then triggers the right protocols at the right time.
Concretely, that means:
On pricing. When the forecast for a future weekend crosses into our “good weather” definition for that property and that season, our pricing moves automatically — not arbitrarily, but inside pre-approved bands. When the forecast moves into our “bad weather” definition with sufficient lead time, the system can also unlock specific promotional levers we’d otherwise hold back.
On operations. Our GMs receive specific advance protocols — staffing, supply ordering, cabin priority assignments, communication templates — triggered automatically when the forecast moves into defined ranges. Nothing is “figure it out Friday afternoon” anymore.
On forecasting and budgeting. The system feeds back into our financial modeling, so we can now answer questions our owners used to consider unanswerable: how much of last August’s revenue performance was weather-driven versus operations-driven? What’s the weather-adjusted comparable for this property year over year? Real decisions get made on that data.
I’ll share what residual benefits we have seen occur in addition to revenue growth.
Cancellation handling time at the front desk has dropped. The single biggest source of front-desk stress on a property — judgment-call cancellation conversations during a bad-weather weekend — has become a structured protocol rather than an improvisation.
Revenue capture in marginal weather windows has improved. This is the quieter win. Properties used to leave revenue on the table during borderline weather, because the front desk would over-credit and over-discount out of caution. The system corrects for that by being precise about which weather actually warrants which response.
We can now answer owner questions we couldn’t before. When an owner asks “why was August soft?”, the answer is no longer a hand-wave. It’s a weather-adjusted analysis with documented variance attribution. That conversation, more than any other, has changed our relationship with our owner partners.
Tourism right now is full of operators chasing the wrong technology investments. AI-generated content. Chatbots. Dynamic pricing models built on competitor scraping. Loyalty programs that nobody is loyal to.
The single highest-leverage data asset in outdoor hospitality is the one nobody has been operationalizing: the seven-day forward forecast for the location of every site on your property.
It was sitting there for the entire history of the industry. We just decided to use it.
The next chapter of professional management in outdoor hospitality is going to be written by the operators who systematize what the industry has historically improvised. Weather is the most obvious example. It will not be the last.
