Exnet transforms multi-domain financial data into execution-ready decisions — continuously, in real time.
Continuous decision loop
Signals → Risk → Scoring → Decisions → Actions
Data Layer
Exnet ingests structured financial data across markets, portfolios, clients, sustainability frameworks, and live event streams — continuously and in real time.
Signals are only one input layer within a broader decision architecture.
Market Data
price (L1/L2) · realised volatility · implied vol surfaces · bid-ask spreads · liquidity depth · cross-asset correlation
Portfolio State
asset allocation · drift vs target · factor exposure · concentration risk · drawdown profile · cash positioning · live monitoring
Client Data
risk tolerance · mandate constraints · time horizon · withdrawal patterns · liquidity needs · behavioural bias signals
Macro Data
interest rate curves · inflation trends · liquidity conditions · credit spreads · risk regime classification
ESG Data
controversy events · carbon intensity (Scope 1–3) · ESG ratings · SFDR / taxonomy alignment · governance risk signals
Live News & Event Feed
breaking company news · earnings events · policy announcements · macro headlines · market-moving incidents · controversy updates
Agent Signals — drift alerts · risk triggers · anomaly detection · behaviour flags
All inputs are standardised, time-aligned, and continuously updated in real time.
All inputs update continuously — not in batches.
Quant + ML
Exnet applies quantitative models and machine learning across all structured inputs — identifying patterns, scoring risk, weighting confidence, and continuously updating decision quality.
Raw inputs
market · portfolio · client · macro · ESG · news · agent
Feature engineering
normalisation · regime detection · anomaly extraction · event classification
Model scoring (ML)
signal ranking · impact estimation · pattern recognition
Confidence weighting
multi-model consensus · historical calibration
Decision intelligence
structured, scored, constraint-aware outputs
Signal Score
87
Confidence
0.94
Risk
elevated
Decision Engine
Decisions are constructed through optimisation — not selected from predefined signals.
Multiple inputs combine into a single optimal action
Inputs
Decision
Filtering
All inputs are filtered through severity, confidence, policy, ESG constraints, and client suitability before reaching execution.
Most signals are discarded. Only the actionable survive.
Live Monitoring
Exnet continuously monitors portfolios across drift, risk exposure, market conditions, ESG events, and client constraints — surfacing high-priority situations in real time.
Monitoring is continuous, not periodic.
When thresholds are breached, actions are triggered automatically.
Hartley Retirement Portfolio
Balanced Growth · 60/40 mandate · £2.4M AUM
Equity allocation drift: +4.2% beyond mandate tolerance
Detected 12 min ago · Impact: £102K AUM at risk
Severity
92
Confidence
0.94
Impact
High
Key Drivers
Equity weight 64.2% vs 60% target (threshold: ±3%)
Elevated market volatility increasing drawdown risk
ESG: carbon exposure above Scope 1–2 limit
Recommendation
Rebalance equity to 60% target · Increase fixed income 2.8% · Raise cash buffer 1.4%
Execution
Decisions are converted into execution-ready workflows — allocations, trades, and rebalancing actions — ready for approval or direct execution.
Routed to workflows, advisors, or execution systems.
Example outputs
Feedback Loop
Execution generates real-world outcome data — performance, timing, behaviour, and client response — continuously refining models, scoring, and future decisions.
Improves signal accuracy, confidence calibration, and timing.
What is captured
Execution outcomes → recalibration
realised performance · slippage vs expected · execution timing
Client response
withdrawals / allocations · behavioural reactions · mandate changes
Market context
regime shift post-decision · volatility evolution
Model update
Model recalibration
signal accuracy tracking · confidence adjustment · feature weighting updates · regime adaptation
Improved decisions
Better outputs
better signal ranking · earlier risk detection · adaptive allocation
Example improvement
Deployment
Embed decision intelligence into existing systems, or deploy a full white-labelled platform.
Embed
API / decision feed layer
Feed live recommendations into CRMs, rebalancers, advisor workstations, and client portals
Preserve existing workflows
Ideal for firms that want Exnet intelligence inside their current stack.
Keeps your workflows — adds intelligence.
White Label
Fully branded advisor and client experience
Exnet Core + monitoring + portfolio decisioning
Workflow, reporting, and client interaction layer
Ideal for firms that want a full decision platform.
Replaces fragmented workflows entirely.
Advisor
Client
Execution
Integrates into existing infrastructure or operates as a full-stack decision platform.
Agents
Exnet agents monitor distinct domains — portfolios, markets, risk, client behaviour, and sustainability events — and produce structured inputs for the Core.
Agents do not make decisions directly. They inform them.
