Claims and loss adjustment
Ucotron measures the event footprint, sorts the claims from the day after, and compares each field before and after. What can't be seen, it says. And every case ends in a signed assessment that a person decides.
Simulation: the footprint of the Córdoba storm of Sep 19, 2026 advances every ten minutes over claims from a demo portfolio, which are sorted into three queues: consistent, inconclusive, and no consistent activity.
What happens today
The loss adjuster has to go to the farm to validate, not to search.
Today
With Ucotron
The day after
With GLM lightning and cloud-top temperature, Ucotron measures the event footprint in high, medium, and low bands. Then it cross-checks it against the claimed fields or homes, so the loss adjuster goes first where it matters most.
Time15:20
Lightning in 10 min58
Consistent
The location falls in high or medium signal within the declared window. The loss adjuster goes here first.
Inconclusive
Low signal, footprint edge, or no image yet. It stays with its reason, waiting for more evidence.
No consistent activity
No signal at the point in that window. It's not a rejection: a person reviews it, with the evidence in front of them.
Verify a hail event
To separate the event's damage from what the field always does, Ucotron compares greenness pixel by pixel at 10 m and subtracts the control from the previous three seasons.
Before · 01/30 · NDVI 0.95
After · 03/01 · NDVI 0.28
Median NDVI of the field0,95 → 0,28Jan 30 vs Mar 1, 2026
With net drop99.1 haof 100.8 ha, 98.3 %
Control 2023–2025+0,001how much the previous seasons moved on those dates
Hectares by net-drop band
Ucotron does not estimate the damage percentage: the loss adjuster defines that.
Hail, water, and lightning
What it doesn't sayThe damage percentage or the loss. With clouds or edges, the pixel stays gray: not measurable.
What it doesn't sayLikely waterlogging is inferred and shown separately from observed water. Usual water doesn't count as an event.
74 of 79cases correctly caught by the flood detector in the golden set
What it doesn't sayGLM detects 70 to 90 % of flashes and doesn't say where a bolt hit the ground. We say “consistent activity at 1.2 km,” not “lightning struck the house.”
The assessment
What was examined, the date of each data point, and what couldn't be measured. It attaches to the claim and holds up to review.
Verification assessment
Verdict
The loss adjuster defines the damage percentage.
Claims recipes
Hail
Runs when a claim comes in on a field.
Compares the field's greenness before and after, pixel by pixel, against its previous seasons and issues a signed assessment.
Hail
Runs when a storm passes with claims filed.
Crosses the event footprint with each claim and sorts them into three queues: consistent, inconclusive and no activity.
Hail
Runs when the satellite passes cloud-free over a claimed field.
Tells the loss adjuster it can now be measured, with the before and after ready.
Hail
Runs when a claim comes in on a policy issued days before the storm.
Flags it so the loss adjuster goes in aware, with the event date and the issue date.
Lightning
Runs when a claim comes in for electrical damage, fire, or dead livestock.
Counts the flashes within 2, 5 and 10 km in the declared hour and issues a signed verdict.
Excess water
Runs when radar or optical sees water where there usually isn't any.
Separates the event's water from the usual pond and counts the days under water.
Rain
Runs when more than 100 mm of rain falls in 24 h over a zone or basin.
Alerts claims with the number of policies and places inside the zone.
Rain
Runs when hail falls or more than 100 mm of rain falls over a city.
Prepares auto, home and condo lines for the wave of claims ahead.
You pick claims you've already settled and send us only the place and date. We build the assessments without seeing the outcome and compare them to the loss adjuster, case by case. No need to share your portfolio.