.jpg)
THE AIDIGITALENGINE RESEARCH STANDARD
A financial research standard built for decisions that move capital
AIDigitalEngine research starts with the financial objective, identifies the evidence that can materially change it, verifies that evidence, tests competing explanations, and defines what would make the conclusion change. The result is research built to support clearer capital allocation, stronger risk control, and better financial decisions.
A conclusion is decision-ready only when the evidence, assumptions, uncertainty, and conditions for changing the view can be traced.
1. OBJECTIVE → 2. EVIDENCE → 3. VERIFICATION → 4. ANALYSIS → 5. CHALLENGE → 6. DECISION → 7. MONITORING
PROFESSIONAL RESEARCH PRINCIPLES
Our process is built around principles that also matter in institutional investment research: diligence, a reasonable basis for conclusions, verifiable evidence, clear communication of material risks, and separation of fact from opinion.
RESEARCH MANDATE & MATERIALITY
Research begins by
deciding what matters
Before collecting data or selecting a model, we define the financial objective, capital involved, decision horizon, constraints, and the conditions that could materially change the outcome. This prevents research from becoming a collection of interesting but financially irrelevant information.
THE RESEARCH MANDATE
1. What needs to be decided?
Define the decision before defining the research:
Allocate Capital · Compare Investments · Evaluate a Company · Reduce Risk · Review a Portfolio · Improve Profitability
A research question should end in a financial decision, not simply an explanation.
2. What is economically at stake?
The required depth of research should reflect the size of the exposure, potential downside, reversibility of the decision, and financial importance of getting it wrong:
Capital Exposure · Downside · Opportunity Cost · Reversibility
3. What limits the decision?
A conclusion can change depending on the time horizon, liquidity requirements, risk limits, financing conditions, mandate restrictions, or operational constraints:
Horizon · Liquidity · Risk Limits · Funding · Mandate
4. What evidence would be strong enough to change the action?
We define in advance what would justify allocating more capital, reducing exposure, delaying the decision, or abandoning the case:
Proceed · Reassess · Reduce · Reject
Without a decision threshold, more research can simply produce more information
The materiality test
Every significant input should pass a simple test before it enters the financial case:
1. Value
Can this information materially change expected value or valuation?
2. Risk
Can it alter downside, probability, liquidity, or exposure?
3. Timing
Can it change when capital should be committed, reduced, or reviewed?
4. Decision
Would a reasonable change in this variable alter what we actually do?
If the answer is no, the information is usually secondary.
1. MANDATE → 2. MATERIALITY → 3. RESEARCH DEPTH
EVIDENCE HIERARCHY & SOURCE INTEGRITY
The closer the evidence is to the source, the more weight it deserves
We rank evidence by origin, timing, reliability, and relevance before it enters the financial case. Primary disclosures and official data carry more weight than commentary, summaries, or model-generated interpretation.
1. Primary evidence: What actually happened?
Regulatory Filings · Audited Statements · Company Disclosures · Central Banks · Exchanges
Primary sources establish the factual record before interpretation begins.
Examples include SEC EDGAR for U.S. filings, FRED for official economic series, and the ECB Data Portal for euro-area statistics. SEC EDGAR provides public access to company filings, FRED aggregates a broad range of economic data, and the ECB Data Portal provides official ECB statistics and metadata.
Decision weight: Highest
2. Professional market data: How the market is pricing the evidence?
Prices · Estimates · Ownership · Transactions · Credit · Market Structure
Professional financial datasets help connect the factual record to valuation, expectations, positioning, and current market conditions.
Decision weight: High
3. Specialist research: What requires deeper domain knowledge?
Sector Research · Technical Expertise · Industry Data · Specialist Interpretation
Some questions require expertise that cannot be reduced to a general database or model. Specialist research is used when the economics of the decision depend on industry-specific or technical knowledge.
Decision weight: Context dependent
4. Analytical models: What the evidence implies?
Valuation · Forecasting · Quantitative Models · Scenario Analysis · AI-Assisted Research
Models help interpret evidence; they do not replace it. Their value depends on the quality of the inputs, assumptions, and validation behind them.
Decision weight: Derived
A sophisticated model cannot repair weak evidence.
4. Commentary & Secondary interpretation: What others think happened?
News · Commentary · Opinion · Summaries · Social Signals
Secondary sources can help identify developments or competing interpretations, but material conclusions should be traced back to stronger evidence whenever possible.
Decision weight: Supporting only
Source integrity test
Every material source should answer five questions:
1. Origin
Where did the information come from?
2. Timestamp
When was it actually available?
3. Version
Has it been revised or restated?
4. Consistency
Do independent sources agree?
4. Materiality
Can it change value, risk, probability, timing, or capital?
If the source cannot be traced, dated, and challenged, its decision weight falls.
What was known at the time matters
Research can become misleading when revised data or later information is treated as if it had been available earlier. For forecasting, backtesting, and historical decision review, we separate point-in-time evidence from information learned afterward.
ANALYTICAL TRIANGULATION
One strong signal is useful.
Independent confirmation is stronger.
We test important conclusions from different angles. Fundamentals may support a thesis while market pricing, positioning or quantitative evidence points the other way. Those disagreements are not ignored, but become part of the decision.
1. Fundamental Evidence
Do the economics support the case?
We test whether the underlying economics support the financial conclusion rather than relying on price movement alone:
Revenue · Margins · Cash Flow · Balance Sheet · Valuation · Returns on Capital
2. Market Evidence
Is the market confirming it?
Market behavior can confirm the thesis, show that it is already priced in, or reveal risk that fundamental analysis has not captured:
Price · Credit · Liquidity · Positioning · Volatility · Cross-Asset Behavior Sheet · Valuation · Returns on Capital
FINANCIAL CONCLUSION
How much confidence does the evidence justify?
3. Quantitative Evidence
Does the conclusion survive different assumptions?
We test how dependent the conclusion is on particular assumptions and what happens when the environment moves outside the central case:
Statistics · Factor Exposure · Sensitivity · Scenario Analysis · Stress Testing
4. Contradictory Evidence
What argues against the conclusion?
We actively look for evidence that weakens the thesis. A conclusion becomes more useful when we understand what could make it wrong:
Alternative Explanation · Missing Evidence · Counter-Signal · Invalidation
The AIDigitalEngine triangulation rule
No material conclusion should rest on a single analytical path
CONVERGENT:
Independent evidence supports the same conclusion:
Fundamentals + Market + Quantitative Evidence
Result: Higher justified confidence.
MIXED:
Some evidence supports the case while other evidence does not:
Support + Contradiction
Result: Reduce confidence, investigate further, or change the structure of the decision.
WEAK:
The conclusion depends mainly on one source, model, or assumption:
Single Path
Result: Insufficient basis for a high-consequence capital decision.
The purpose is not agreement for its own sake.
Triangulation helps distinguish a robust financial case from one that appears convincing only because the same assumption is being repeated through different models or sources:
Higher confidence
when independent evidence converges.
Lower decision error
when contradictions are identified before capital moves.
Clearer risk
when the client knows which part of the thesis remains uncertain.
Confidence should rise because independent evidence converges—not because one model sounds certain.
When the evidence conflicts, we lower confidence, identify where the disagreement comes from, and decide whether more research can settle it before the conclusion influences capital.
SCENARIO, RISK & INVALIDATION
A strong research case must show
what can break it
We do not rely on one forecast. We test how the conclusion shifts under different conditions, locate where the downside sits, and define the evidence that would make the original view no longer defensible.
1. Base Case
What is the most defensible outcome today?
Built from the evidence currently available:
Current Evidence · Central Assumptions · Expected Outcome
2. Upside Case
What has to improve for the outcome to become better?
Built from the evidence currently available:
Current Evidence · Central Assumptions · Expected Outcome
3. Downside Case
What happens if key assumptions weaken?
Measure how lower growth, weaker margins, tighter liquidity, higher rates, or other pressures affect the financial case:
Weakness · Exposure · Loss
4. Stress Case
What happens outside normal expectations?
Test conditions that are unlikely but financially important:
Liquidity Shock · Credit Stress · Volatility · Tail Risk
5. Invalidation
What evidence would make us change the conclusion?
A conclusion should have a condition for changing it:
Thesis Break · Risk Limit · New Evidence · Structural Change
WHAT WE MEASURE:
Probability
How likely is each scenario?
Magnitude
How large is the financial impact?
Liquidity
Can the position be changed if conditions deteriorate?
Reversibility
How difficult is it to correct the decision?
Research is stronger when the downside is defined before the decision is made.
The objective is not to eliminate uncertainty. It is to understand how much uncertainty the capital decision can reasonably absorb.
SPECIALIST & INSTITUTIONAL RESEARCH ACCESS
We use the research environment the financial problem requires
Some decisions can be answered from primary filings and public market data. Others need institutional datasets, private-market information, specialist sector research, quantitative expertise or the client’s own financial data.
ACCESS LAYER
1. Professional Market Infrastructure:
Market Data · Estimates · Ownership · Transactions · Research · Credit
2. Private Capital Research:
Private Companies · Funds · Deals · Investors · Credit · Valuation
3. Domain Expertise:
Industry · Technical · Regulatory · Accounting · Quantitative
4. Qualitative Research Depth:
Transcripts · Expert Interviews · Research Libraries · Internal Documents
WHEN IT IS NEEDED
When public sources do not provide enough depth, consistency, history, or market context for the decision.
When the decision involves private companies, venture capital, private equity, transactions, fundraising, acquisition targets, or private credit.
When the economics of the decision depend on knowledge that cannot be reliably inferred from general market data.
Examples:
-
energy and commodities
-
banking and credit
-
technology infrastructure
-
insurance
-
real estate
-
pharmaceuticals
-
industrial supply chains
-
accounting complexity
-
quantitative portfolio risk
When the question depends on management language, competitive dynamics, customer behavior, supplier conditions, industry knowledge, or qualitative information that does not appear cleanly in financial statements.
WHAT IT ADDS
Standardized financials, consensus expectations, private-company coverage, market data, ownership, transactions, research, and professional analytical workflows.
Comparable transactions, investor activity, deal history, fund information, financing data, and private-company intelligence that public-market databases may not capture adequately.
Specialist interpretation of variables that may materially alter valuation, risk, operating assumptions, or the probability of the outcome.
A model should not substitute for expertise it does not possess.
Deeper context around what is changing before the effect becomes fully visible in reported financial results.
Core AIDigitalEngine principle
Access alone is not the advantage
Professional platforms can provide enormous amounts of information. The research advantage comes from knowing which source is appropriate, which evidence is material, what must be independently verified, and how the information changes the financial decision.
Selection:
We use only the data the problem requires.
Verification:
We trace important information back to its source.
Integration:
We combine market, financial, quantitative, and specialist evidence.
Decision:
We translate the evidence into a clear financial consequence.
The research stack should expand only when the financial consequence justifies the additional depth.
AIDigitalEngine does not add data for appearance. We add sources, specialists, and analytical depth when they materially improve the basis for the decision.
DECISION TRACEABILITY & DELIVERY STANDARD
Every important conclusion should be traceable back to its evidence
A financial conclusion should not arrive as a black box. We show the evidence behind it, the assumptions applied, the method used, the risks identified, and the conditions that would require the decision to be reviewed.
1. Evidence → 2. Assumption → 3. Method → 4. Scenario → 5. Risk → → 6. Decision
1. Evidence
What supports the conclusion?
Primary data, market information, specialist research, or client-provided evidence.
2. Assumption
What are we assuming?
We make explicit the variables that are not directly observable or certain.
3. Method
How was the evidence analyzed?
We use the method appropriate to the financial problem.
4. Scenario
What changes if the assumptions change?
We show the central, upside, downside, and stress cases.
5. Risk
What could materially weaken the case?
We identify downside, uncertainty, concentration, liquidity, and invalidation conditions.
6. Decision
What should the evidence change?
If the research does not change the decision, its financial relevance should be questioned.
EVERY CLIENT OUTPUT SHOULD MAKE THESE CLEAR:
Facts:
What is directly supported by evidence?
Assumptions:
What depends on judgment or estimates?
Confidence:
How strong is the supporting evidence?
Downside:
What can materially go wrong?
Invalidation:
What would make the conclusion no longer defensible?
Next Review:
What should be monitored after the decision?
Delivered for use, not for storage
The final output should be understandable enough to use immediately and detailed enough to audit later. Depending on the assignment, that may take the form of a financial intelligence report, decision framework, custom system, or recurring intelligence process:
Financial Intelligence Reports → Financial Systems & Playbooks → Custom Financial Intelligence Solutions → Recurring Intelligence
The client should be able to reconstruct the decision without guessing how we arrived there.
That means separating facts from assumptions, showing where uncertainty remains, defining what would change the conclusion, and leaving a clear record of how the evidence became a financial decision.
-
Overview
-
Mandate
-
Evidence
Bring us the decision that needs
better evidence
Tell us what you are trying to understand, compare, protect, improve, or decide. We will determine the research depth, evidence, analytical methods, and specialist input required before the decision is made.