Today, we're launching our first commercial offering, targeted at outdoor contractors who need to know the weather and site conditions before they go. What most site managers do for landscaping and roofing is just to check their phone's weather app for the various locations they need to dispatch crews, and try to figure out if any jobs should be cancelled or postponed. One owner told me he wakes up at 4am every day, because if he lets them know after 6am, he has to pay them even if they can't work.
Prelay monitors the entire upcoming schedule of sites and determine suitability of 62 activities across 12 trades. We're using ensemble models for weather, and a custom ground-wetness model for soil conditions. Users can also configure weather-triggered notifications so they reschedule work as early as possible.
As of today, Prelay is available in the US, and people can check it out at https://prelay.lwnn.news - I'm super interested in hearing from anyone who works in weather-sensitive businesses or has built similar software in this (or another) niche.
One question we've gone back and forth on is where the line should be drawn between providing raw weather data and recommending concrete actions (something like, 'there's a 35% chance of rain' vs 'you should consider postponing this job'). I'm curious what HN thinks of that question, and what you'd suggest.