Capability / 01 / Match

Match

Two-sided pairing, fully reasoned. Ranked output with sourced rationale and evidence on every score.

platform / 01 / matchRANKED · 3 LINESCLIENTMATCHING ENGINERANKED CARRIERSAcme Mfg.Manufacturing · CAProperty$8M TIVWorkers' Comp$2.4M payrollCommercial Auto12 vehicles+ 4 MORE LINESMatching engine6 DIMENSIONSCarrier 0270.942Carrier 1420.813Carrier 0580.73Carrier 0270.892Carrier 2110.783Carrier 1420.71Carrier 1420.862Carrier 0270.793Carrier 0580.68+ 44 MORE EVALUATED PER LINEEVERY CARRIER SCORED · RANKED OPTIONS WITH REASONING PER LINE

Your matching process cannot explain itself

01

A single score hides every tradeoff

You collapsed a multi-dimensional analysis into one number. Your team cannot see the tradeoffs. Your clients cannot understand the output.

02

Regulators ask, you improvise

A compliance officer asks why Client X was placed with Carrier Y. Your team reverse-engineers the reasoning from memory. The audit trail is a person, not a system.

03

Domain knowledge leaves when people do

Your best people carry the real matching logic in their heads. When they leave, years of pattern knowledge leave with them. You cannot scale what you cannot encode.

How Match fits into your matching work

Input

A subject and a candidate set

A subject (a deal, an account, a candidate brief, a property) and a candidate pool (carriers, sellers, candidates, books). Plus your matching criteria: hard rules, scoring dimensions, exclusion logic.

Output

Ranked output, sourced rationale per pair

Each candidate scored across the dimensions your firm cares about, with sourced rationale per pair. Categorical scoring, not opaque numerics. Exclusions visible, reasoning attached, output ready to defend.

Configured

Your scoring, your appetite, your exclusions

Configured to your firm's dimensions, gates, and house standards. The platform learns from your historical determinations, not a generic baseline.

Integration

Reads from your stack, routes to your tools

Direct APIs into AMS, CRM, and candidate databases. MCP servers, partner feeds, and document parsing where systems aren't API-first. Routed output lands wherever your team already works.

One API call. Ranked output with sourced rationale

What you send to the platform, and what you get back. Categorical scoring across six dimensions, evidence per dimension, audit trail.

request.pyRequest
import nexio

client = nexio.Client(
  api_key="nx_live_k8s2..."
)

# Evaluate a match in one call
result = client.evaluate(
  profile={
    "entity": "acme-manufacturing",
    "state": "CA",
    "line": "professional_liability",
    "revenue": 12_000_000
  },
  pool="carriers:all"
)

Match evaluation

Acme Manufacturing

TIER 1

Assessment: Strong fit for CA professional liability

Dimension
L1
L2
L3
L4
Appetite fit
Ideal
Coverage breadth
Solid
Financial strength
Superior
Pricing
Competitive
Match likelihood
High
Service quality
Good
Assessed: Feb 2026Confidence: HIGH6/6 dimensions at L1–L2

Watch the platform work

The three-stage platform described above, running live against sample data. No sign-up required.

nexio evaluate --interactive

Entity Profile

Entity:acme-manufacturing
State:CA
Line:professional_liability
Revenue:$12,000,000
Employees:145
Prior Carrier:Carrier W
Claims (3yr):2 closed, 0 open
Effective:2026-07-01
Pool:carriers:all

Press Run Evaluation to start

Book a demo

See Match in action

Walk through example deployments and how Match maps to your firm's pairing work. Inspect every score, exclusion, and reasoning step.