Optimisation improves efficiency, Allocation determines growth

SignalOS™ connects BI data (Finance, CRM, Data Science) to weekly Marketing allocation decisions.

Not another BI dashboard. (You already have Tableau/Looker.)
Not financial modeling. (Finance consultants do this.)
Not a media buying tool. (Agencies do this.)

A decision-making operating system for €20–200M portfolios managing bonus economics, player quality, and multi-market allocation.

Built from managing €200M+ marketing budgets. Battle-tested across 6 regulated markets.

Most operators already have the data.

But allocation decisions are still made using platform dashboards that don’t reflect total cost, customer quality, or forward value.

That’s where inefficiency creeps in.

SignalOS™ builds the missing layer between insight and action.

It takes the metrics that actually drive profitability and turns them into a structured weekly decision system.

Define what success looks like before capital moves.

You set:

• Allowable CPAs by product vertical (Sports €120, Casino €80, Live €100)
• Player quality thresholds (multi-deposit % ≥40%, 7-day active % ≥50%)
• Product mix targets (Sports 60%, Casino 35%, Live 5%)
• Margin gates (minimum NGP/CPA ratios by market maturity)
• Event-driven ranges (how much CPA can expand during World Cup, Champions League)

INTENT acts as a constitution — every allocation decision is tested against these strategic guardrails.

Collect 8 performance signals across multiple dimensions:

DEMAND signals:
New customer %, search volume trend, competitor share-of-voice

COST signals:
Total CPA (media + bonus), CPM trend, frequency pressure

CONVERSION signals:
Registration-to-FTD rate, payment success rate, form abandonment

RETENTION signals:
Multi-deposit %, 7-day active %, 30-day active %

MARGIN signals:
30-day NGP, NGP/CPA ratio, bonus cost per FTD

COMPETITION signals:
Competitor creative frequency, bonus aggression, auction pressure

EFFICIENCY signals:
Click-through rate, creative fatigue index, audience saturation

CONTEXT signals:
Event tier classification, regulatory changes, seasonality

Each signal receives a directional assessment: POSITIVE (green), NEUTRAL (amber), NEGATIVE (red).

The framework aggregates signal strength and generates recommendations.

Logic:

• If 6+ signals positive + margin gate PASS → 🚀 ACCELERATE
• If 6+ signals negative OR margin gate FAIL → 🛡️ DEFEND
• If signals mixed or inconclusive → ⏸️ HOLD

Special rules:

• Margin gate failures override all other signals (profitability is absolute)
• CREATION vs HARVEST classification influences allocation aggressiveness
• Event tier classification expands or contracts CPA tolerance windows

Every recommendation is logged with:

• What signals drove the decision
• What capital movement was made (€XXk from Channel A to Channel B)
• What outcome validation will occur (check results in 2 weeks)

Did the recommendation work?

Two weeks after a capital move:

• Compare predicted outcome vs actual outcome
• If recommendation was ACCELERATE → did volume increase without margin collapse?
• If recommendation was DEFEND → did margin stabilise?
• If prediction failed → why? (Signal was wrong? External shock? Execution issue?)

VERITAS creates a feedback loop: bad recommendations get investigated, good recommendations get reinforced, signal quality improves over time.