Using LambLens
Understanding evidence and uncertainty
How LambLens distinguishes evidence, user-provided information, assumptions, interpretation, challenge and uncertainty.
Last reviewed: 2026-08-28
LambLens works best when the basis of a conclusion is visible. This guide explains how LambLens distinguishes evidence, user-provided information, assumptions, interpretation, challenge and uncertainty.
The evidence chain
When LambLens investigates, information flows through connected stages:
- Source — where information comes from (primary data, published research, company information, user context)
- Evidence — what is actually supported by that source (a specific finding or data point)
- Claim — the conclusion or assertion being made based on evidence
- Assumption — what has to be true for the claim to hold (dependencies that may or may not be tested)
- Implication — what follows if the assumption or claim is correct
- Validation state — what has been checked and what remains unresolved
This is not a mandatory linear sequence that every investigation must follow. Rather, it illustrates how evidence connects to reasoning, where assumptions enter, and where validation is necessary. Understanding these relationships helps you inspect whether the developing view is as robust as it appears.
Why evidence is separated from interpretation
A decision can look clearer than it really is when facts, assumptions and conclusions are blended together. LambLens separates them so a user can inspect what is known, what has been supplied, what is inferred, what is qualified and what remains unresolved.
This distinction is central to trust. A conclusion may be useful, but it should not be mistaken for the evidence that supports it.
Source fact
A source fact is information directly supported by an identifiable source. Source quality and applicability still matter.
A source fact is not automatically complete, current, applicable or decisive. A strong source may support one narrow claim without settling the whole decision.
User-provided information
User-provided information is context or factual information supplied by the person or organisation using LambLens. It may include internal cost, constraints, goals, timing, existing contracts, preferences or organisational circumstances.
LambLens may need to treat supplied information differently depending on whether it is known, asserted, estimated or uncertain. Important supplied context can materially change the investigation.
Assumption
An assumption is something the case depends on that has not yet been established as fact. Some assumptions are harmless. Others are material because changing them would change the Current LambLens View, Recommendation or next action.
LambLens should identify material assumptions, make them visible, test them where possible and show when the view depends heavily on them.
LambLens interpretation
A LambLens interpretation is LambLens's reasoned reading of what evidence means in the context of the investigation.
Interpretation is not the same thing as a source fact. A source may say one thing; LambLens may infer what that means for a particular decision only after considering context, applicability, alternatives and uncertainty.
Derived comparison or calculation
LambLens may derive comparisons, estimates, scenarios or calculations from underlying evidence and assumptions. These can be useful, but they should remain distinguishable from raw source facts.
A derived number is not independently sourced merely because it is precise. It depends on the inputs and assumptions used to produce it.
Qualification
A qualification limits how broadly or confidently evidence can be applied. Evidence may be older than ideal, come from a different geography, measure a proxy, cover a different population or depend on an incomplete sample.
Qualifications do not make evidence useless. They explain its proper scope.
Challenge
Challenge is evidence or reasoning that tests the current case rather than simply supporting it. LambLens should look for contradictory evidence, downside scenarios, alternative explanations and conditions under which another option becomes stronger.
A good investigation does not silently choose whichever evidence best supports the first apparent answer.
Uncertainty
Uncertainty is something material that cannot currently be resolved from available evidence. LambLens should expose uncertainty rather than quietly convert it into false confidence.
Not every uncertainty must be resolved before a useful conclusion can be formed. The important question is whether the uncertainty could materially affect the conclusion or next action.
Evidence authority
Different sources can carry different evidential weight. Primary sources, official sources, regulators, academic work, industry research, company disclosures and reputable media may all be useful in different circumstances.
There is no single universal hierarchy that works for every claim. Authority depends partly on what the evidence is being used to support.
Applicability
Strong evidence can still be weak evidence for a particular claim if it does not match the geography, timeframe, population, product, decision context or requirement in the investigation.
LambLens should consider whether evidence actually applies to the question being decided, not only whether the evidence is credible in general.
Freshness
Evidence may become less useful as conditions change. Prices, regulation, market conditions, company disclosures and technology capability can all move over time.
LambLens should consider how current the evidence is relative to the question. This does not imply automated monitoring of every source.
Contradiction
Two sources may disagree, measure different things, use different scopes, apply to different periods or create an unresolved contradiction.
When contradiction is material, LambLens should make it visible and explain its effect on the view rather than ignoring it.
Evidence can change LambLens's view
An investigation may begin with an initial view. New evidence can change an assumption, alter LambLens's interpretation or show that a different alternative is stronger. The Current LambLens View should update when the case genuinely changes.
Evidence can change LambLens's view. A change should follow new evidence, context or constraints, not an arbitrary shift in wording.
For a published example of evidence changing a conclusion, see LambLens Research.
What remains uncertain
A useful investigation should identify material unresolved uncertainty. Not every unknown matters equally.
LambLens should focus attention on uncertainty that could affect the conclusion, Recommendation or next action.
What would change LambLens's view
What would change LambLens's view means the specific evidence, changed assumptions, new constraints or test results that could materially alter the current conclusion.
This helps users see whether the case is robust, what to investigate next and when a decision should wait for more information.
Fact / interpretation / assumption / uncertainty
- Fact: supported by evidence, while still subject to scope, freshness and applicability.
- Interpretation: what LambLens reasons that evidence means in the decision context.
- Assumption: what must currently be treated as true for the case to hold.
- Uncertainty: what remains unresolved and may affect the conclusion or next action.
Source count
More sources do not automatically make evidence stronger. Source count can be useful context, but it is not a substitute for authority, applicability, freshness, relevance and consistency.
Related pages
Next: How LambLens reasons and where responsibility remains.