Tessera HAIL is our hail impact sensor. It counts every strike, measures the kinetic energy each one delivers, records how densely they landed and how long the event ran, and stamps the whole thing to the second. Radar infers hail from a beam sweeping thousands of feet overhead. We measure it where it lands.
PATENTEDMEASURED, NOT INFERREDSOLAR POWERED · NO WIRINGTIMESTAMP TO THE SECOND
Mount detailsSheet 3C · Scale NTS
TAKES A HIT, KEEPS COUNTING. ROOF, POLE OR TRACKER: ONE RECORD PER STORM.
What it actually measures
Five things an instrument knows and a radar model cannot
These five measurements are the raw record. Everything downstream — a parametric trigger, a claim file, a hazard score — is built from them. They are measured at the property, by hardware standing in the storm, which is why they can be stated rather than estimated.
COUNTEvery impact, recorded one at a time
JOULESImpact energy, per stone and aggregated
PER m²Impact concentration across the panel
DURATIONStart, peak and end of the event
HH:MM:SSTimestamped to the second
Impact energy matters most. A roof, a windshield or a solar module fails when enough kinetic energy arrives, not when a stone reaches a nominal diameter. Energy is what the panel feels, so energy is what we report first.
Concentration is the one nobody asks for. Two storms with an identical peak stone leave very different roofs behind. Fifty strikes per square meter and five are not the same event, and only an instrument on site can tell them apart.
Duration and timestamp settle the arguments. A record that says the event began at 22:38:14, peaked at 22:41:07 and ended at 22:49:02 takes the date of loss out of the negotiation. Temperature, relative humidity and tilt angle are recorded alongside it, so the event is read in the conditions it actually happened in rather than in conditions borrowed from a station down the road.
KNOWS THE WEATHER IT STANDS IN · REPORTS WHAT IT FELT, NOT WHAT A MODEL EXPECTED IT TO FEEL.
The honest number
Maximum stone size is a floor, not a guess.
Every buyer asks for a diameter, so here is exactly what ours means. Tessera measures every stone that strikes it. The largest stone in a storm may have come down in the parking lot, on the far side of the building or in a field a kilometer away, and no instrument can measure a stone that missed it.
So the maximum diameter on every report is the largest stone the instrument actually caught, derived from the energy it delivered. It is never inflated. It is a floor: the storm was at least this severe.
And the floor sits close to the truth. Across our multi-sensor sites, neighboring sensors differ by 15% on average, so a single Tessera catches most of what a storm drops on its site. The event report builds on that: it models outward from the measured floor, using every data point around the site, and adds the most likely size beside the one that was caught, labeled as modeled. No sensor on site? Hailios Fusion AI, trained on Tessera measurements, answers for your address.
A remote estimate can land on either side of the truth without saying which. Ours starts from a measured number that reality can only exceed, then tells you what the storm most likely was. Energy and concentration, the things that break roofs, arrive in the same record. That is a diameter you can write into a policy and stand behind in a claim.
WHAT WE CAUGHT IS THE FLOOR. WHAT WE MODEL IS LABELED. NEITHER IS INFLATED.
Validation
Tested in the lab. Proven in the field.
In the lab. At the IBS Hail Lab in Linz, Austria and at Koontz Labs, ice launchers fire stones of known size and mass at the instrument, at known velocity, so the energy arriving at the surface is a controlled quantity rather than an argument. What Tessera reported was compared against what the launcher delivered, shot after shot. Laboratory ice is how a field instrument earns the right to publish a joule figure.
In the field. Across our multi-sensor sites, neighboring sensors differ by 15% on average: independent instruments, the same storm, close agreement.
Over time. Six years of measured record, on three continents, is what Hailios Fusion AI is trained on and what every ground-trained answer is scored against.
Two Tesseras on the target board · IBS Hail Lab, Linz, AustriaTHE LASER MARKS WHERE THE STONE WILL LAND.
A test stone, before the shotKNOWN SIZE. KNOWN MASS. NO ADJECTIVES.Weighed before it’s fired · Koontz LabsTHE ENERGY IS SET BEFORE THE STONE LEAVES THE BARREL.The launcher · IBS Hail LabCOMPRESSED AIR, A MEASURED BARREL, A STONE AT A KNOWN SPEED.On the test bench · Koontz LabsWHAT IT REPORTED, CHECKED AGAINST WHAT WAS FIRED.
THE LAUNCHER KNOWS EXACTLY WHAT IT THREW. THAT’S THE ONLY WAY TO GRADE THE ANSWER.
The instrument
One tile, three mounts, no wiring
Primary measurements
Impact count, impact energy in joules per strike and aggregated across the event, impact concentration per square meter, and the duration of the event from first strike to last.
Time base
Every impact is timestamped to the second. Date of loss stops being a position and becomes a line in a record.
Stone diameter
Derived from measured impact energy and reported as a measured floor, never inflated. The event report adds the most likely size beside it, labeled as modeled.
Supporting conditions
Temperature, relative humidity and the instrument’s tilt angle are recorded alongside the impact record, so every event is read in the conditions it actually happened in rather than in conditions assumed from a nearby station.
Mounting
Roof, pole, or solar tracker. The roof mount is peel-and-stick: adhesive pads onto the membrane, no penetration, no wiring, no site engineer.
Sensing surface
The solar panel itself — our patented use of a solar panel as the impact collection surface. Ice strikes a known, instrumented surface, and the instrument measures the strike rather than inferring it from a distance.
Where the data goes
Into Mosaic, the platform, and out as three answers: Mosaic Risk for the hazard, Reports for the event, Alerts for what is coming.
Deployed
Operating today across North America, Europe and Australia, and deployable anywhere there’s a cellular signal.
EVERY FIGURE THIS INSTRUMENT PUBLISHES IS ONE IT MEASURED. THE ONES IT DIDN’T, IT DOESN’T PUBLISH.
The solar panel is the impact surfaceOUR PATENTED USE OF A SOLAR PANEL AS THE IMPACT COLLECTION SURFACE. EVERY NUMBER WE PUBLISH STARTS HERE.
ROOF. POLE. TRACKER.
KNOWS THE WEATHER IT STANDS IN.
COUNTS. WEIGHS. TIMESTAMPS.
A MEASURED FLOOR. NEVER INFLATED.
Where it goes
Three mounts, because risk does not only live on roofs
Roof
PEEL-AND-STICK
Adhesive pads straight onto the membrane — no penetration, no fixings, no wiring. The instrument sits on the insured surface itself.
Pole
OPEN GROUND
For exposure that is not under a roof at all: vehicle staging yards, equipment compounds, orchards and open fields. One tile, one pole, no trenching and no power run.
Solar tracker
MOVES WITH THE ARRAY
Mounted on the tracker, so the instrument sees the same sky at the same angle as the modules it reports on. Stow decisions come off measurement taken in the plane of the array.
Deployed on a flat commercial roofWHEREVER THE STORM COMES FROM, THE INSTRUMENT IS ALREADY HERE.
Installed once, and then simply present: recording the events that happen, and recording that nothing happened on the days nothing did. A season of quiet is data too — it is the part of the record that makes the loud day credible.
Sensor or ground-trained
Absolute truth where it matters. Ground-trained everywhere else.
Not every asset needs its own instrument. The ones that do are the important, expensive ones, where a single storm can decide a large loss.
MEASURED
Tessera on site
For the assets that need the absolute record. Every stone that strikes the instrument, counted and measured at the asset itself.
GROUND-TRAINED
Ground-trained by Hailios Fusion AI
For everything else, or wherever price decides. Hailios Fusion AI, trained on Tessera measurements, gives each address a ground-trained answer without a sensor on site. It is less exact than an instrument on the asset, and we don’t pretend otherwise: its skill is scored against the ground, and every figure is labeled ground-trained.
One instrument settles one claim. Years of their measurements train Hailios Fusion AI, and together they become something no forecast archive can be: a record of hail that was measured, or ground-trained on what was measured, rather than inferred from radar. We operate today across North America, Europe and Australia, and the sensor is easy and inexpensive to deploy and needs no calibration, so it goes anywhere there’s a cellular signal.
MeasuredGround-trained
MEASURED WHERE WE ARE, GROUND-TRAINED WHERE WE AREN’T — AND LABELED ACCORDINGLY.
HAIL is the Tessera instrument. See all of Tessera, including the WIND add-on, which brings validated partner instrumentation onto the same site record.
Mosaic · the platform
What the panel felt becomes three answers
The measurement is the beginning, not the product. Every impact this instrument records goes into Mosaic and comes back out as risk, as evidence, or as a warning.
