From the file
Wind speed contour map vs. hail swath map for claims triage

Every CAT lead has pulled both products up on the same screen at some point and watched them disagree. The wind contour shows a long gust footprint running northeast. The hail swath shows a tighter, lumpier band of MESH values thirty miles off to the side. The claims queue doesn't care which model is right. It needs a priority order by lunchtime.
What a wind speed contour map shows
A wind contour map is built from modeled gust estimates, usually interpolated from station observations, reanalysis data, or a vendor's own wind field model. It answers one question: how fast was the air moving at a given point during the event. That's useful for triggering policy deductibles, for flagging PCS-code exposure, and for a first pass at which zip codes to stage adjusters in.
What it doesn't tell you is whether a roof failed. Wind speed is an input to damage, not a measurement of it. A 70 mph gust against a twenty-year-old three-tab roof and the same gust against a roof installed last spring produce different claims, and the contour line draws no distinction between them. Use it to plan where to stage crews. It can't tell you which claims need daylight first.
What a hail swath map shows
A hail swath map, usually derived from radar-estimated MESH (Maximum Estimated Size of Hail) or from hail reports, traces the path of the hail core across the ground. It's narrower and more erratic than a wind contour because hail cores wobble, skip, and sometimes split. Inside that swath, stone size estimates vary block to block, not just county to county.
This is the layer most CAT teams already use to draw their initial "in scope" boundary for a hail event. It's a reasonable first filter. The problem is resolution. A swath polygon covering six zip codes tells you the event touched that area. It doesn't tell you which of the four thousand policies inside it have bruised shingles or a displaced ridge cap, versus which ones sit on the shoulder of the core and took nothing worse than some granule loss.
Why triage needs both, in the right order
Run the wind contour first to scope the catastrophe and set initial reserves. Run the hail swath next to narrow the geography. Then you're left with the part neither product solves: a priority order inside that footprint, address by address, before the first truck rolls.
That's the gap between a modeled hazard layer and an actual damage layer. Contours and swaths are both inference. They tell you where the storm could have done damage based on physics and radar return. They don't tell you where a roof shows it. Two houses on the same street, same construction, same age, can land in very different places once you look at them.
Where per-address imagery fits
This is where a flyover or VHR satellite pass over the swath earns its keep. Instead of modeling probable damage from atmospheric data, it photographs the roofs directly after the event and flags visible bruising, missing shingles, and displaced ridge caps per address. That turns a six-zip-code polygon into a numbered list: these three hundred addresses show visible roof damage, these eleven hundred look untouched, go look at the rest in person.
Hail & Wind Damage runs that pass across the whole swath and delivers the numbered memo the morning after the event, so the dispatch order for day one reflects what the roofs show instead of where the contour lines happen to fall.
Neither the wind contour nor the hail swath map was built to answer "which address do I send someone to first." They were built to answer "did this event happen, and roughly where." For the queue-ordering question, you need a layer that actually looked at the roof.
If your queue is still sorted by swath polygon alone, it's worth seeing what a per-address pass adds before the first adjuster goes out.