Squall · how it works

Proprietary instrument

SQUALL

A squall arrives fast, over water that looked calm a minute ago.

The tools get better, so does the pressure on people

Detection software catches some of it. The rest still lands on a person, with a deadline, deciding in seconds whether to act.

An agent, not a questionnaire

Built on your world, not an invented one

The usual way

Tests people on invented scenarios.

The Squall agent

Uses your own fraud and claims patterns, inside the working week.

Step one

What it works with

Fraud and claims patterns, anonymisedTraining platformEmail, for delivery onlyTeam structure
Squall agent

Opt-in and anonymised where people are involved.

Step two

How it investigates

build12 scenarios built from last quarter's fraud patterns
deliversent inside the existing training platform to 6 teams
scoreaccuracy 74%. Wrong and certain in 11% of answers.
flagclaims team B is most confident when it is wrong
adaptnew attack pattern added: a cloned-voice request
trackconfidence matched reality in 3 of 6 teams, up from 1

Every line is something a person can check afterwards.

Step three

What it produces

TeamAccuracyConfidence
Claims A81%Matches
Claims B69%Overconfident
Payments77%Matches
Customer service72%Underconfident

Accuracy and calibration by team.

How it is held to account

Rules, not vibes

01

Wrong and certain is what costs money

Each answer is scored for accuracy and for whether confidence matched reality.

02

Escalating can be the right answer

Where a scenario is genuinely uncertain, guessing is scored as the error.

03

Teams, not individuals

Results are totals by team, with nobody named.

The agent is held to the same questions it asks of everyone else.

Configured to you

Set up around your organisation

01

Your cases

Scenarios built from your own claims and payment types.

02

Your channels

Delivered where people already work and train.

03

Your teams

Grouped the way your operation is organised.

How an engagement runs

From first conversation to a working agent

1

Scope

Which teams, which fraud types.

2

Build

Scenarios from your own patterns.

3

Run

Inside the working week.

4

Track

Judgement over time, as new attacks appear.

A typical shape. The sequence and timings are set with you.

One platform

Squall does not work alone

KeelSets the policy and the gates for every use case.
SentinelQuestions what each agent can see, change and cause.
SoundingsSees what is actually running across the organisation.
BearingProves it to the supervisor, obligation by obligation.
The human layerHALO, Squall and Ensign measure what no system log can.

See the whole system on the Solutions page.

That's Squall

See what your own judgement actually looks like under a clock.

A demonstration of the method is one click away. The deployed Squall is built for your environment and configured to your regulator.

Discuss a deployment