What is event-driven tasking in post-earthquake satellite imagery?

If you've worked a tasking request before, you know the phrase gets thrown around loosely. Vendors say "we can task for you" and leave it at that. For an EOC trying to get eyes on a damage zone within hours of a quake, the distinction between event-driven tasking and everything else matters more than the sales pitch suggests.
Event-driven tasking means a satellite's collection schedule gets reprioritized because something happened, not because it was already on a rotation. Most commercial imaging satellites run standing collection plans: they pass over cities, farmland, coastlines on a cadence set months in advance, filling archives that get sold later. An earthquake doesn't care about that schedule. Event-driven tasking is the mechanism that interrupts it, pushing your affected area to the front of the queue the moment the ground stops shaking.
Tasking versus pulling from the archive
Someone on your GIS desk can pull last month's imagery of the affected neighborhoods from the archive in minutes. That imagery shows the buildings before the quake, not after. For search-and-rescue triage, pre-event imagery works as a baseline for comparison, not as the thing you dispatch teams against.
Tasking is the request that gets a satellite pointed at your footprint on its next usable pass, weather and orbit geometry permitting. "Event-driven" just specifies why that request jumped the line: a seismic trigger, not a subscription schedule. In practice this is what separates a dated, structure-level read of the damage zone from a street survey that takes field teams days to complete on foot.
What changes about the timeline once tasking is event-driven
Nobody can promise you a fixed number of hours. Satellite revisit depends on orbital mechanics, cloud cover, and how many other requests are competing for the same sensor at the same time. What event-driven tasking changes is priority: your earthquake footprint stops waiting behind routine agricultural or urban monitoring collects and gets scheduled as soon as a pass is physically possible.
That's a different posture than "we'll get to it." It's the difference between a tasking order that sits in a general queue and one flagged for immediate processing because a disaster triggered it. For very-high-resolution imagery at 0.5 meter ground sample distance or better, the kind where you can actually tell a pancaked roof from an intact one, this prioritization is what makes the collect usable for triage instead of arriving after the triage window has already closed.
What a coordinator should ask a provider
When you're on the phone setting up tasking, ask three things plainly: what counts as the trigger for prioritization, what the realistic next-pass window looks like given current orbital coverage over your region, and what resolution and band the collect will come back in. A provider who can't answer the third question with a specific number isn't giving you tasking, they're giving you a promise.
This is also where the gap between tasking and classification gets confused. Tasking gets you the picture. It doesn't grade the buildings in it. A raw VHR pass over a collapsed district still requires someone, or something, to go structure by structure and sort what's standing from what isn't. That's a separate step, and it's worth asking your provider whether it's included or whether you're on your own once the pixels land.
Satellite tasking after a disaster only earns its keep if the output gets turned into something a search-and-rescue sector chief can act on. Building Damage Classification takes that post-tasking pass and turns it into per-structure damage grades, so the imagery cadence you fought to prioritize feeds the triage decision instead of sitting in a GIS layer waiting for someone to eyeball it.
If your EOC is setting up tasking protocols before the next event instead of during it, it's worth looking at what a structure-by-structure read of that imagery could add to your plan.