A positive outcome is not automatically a proven win. The same upward movement can come from a controlled test, a seasonal before-and-after comparison or an anecdotal observation. Those findings should not travel through the organisation with equal certainty.
A small shared vocabulary makes decision records more honest and more reusable.
Marketing evidence levels describe how a result was learned, while the outcome describes what direction it moved. Keeping those labels separate prevents weak evidence from becoming a strong claim over time.
Not measured
No reliable performance result is available. This is a valid label for brand, legal, strategic or capacity decisions and for changes where measurement was not practical. Record the decision basis rather than inventing a metric.
Observed
A directional movement was noticed after or during the change, but there was no reliable comparison design. Observations are useful for triage and hypothesis generation, not causal claims.
Before and after
A defined period after the change was compared with a historical baseline. Record window choice, traffic mix, seasonality and concurrent changes. This supports a more structured directional conclusion.
Controlled experiment
Participants or exposures were assigned to comparable conditions and analysed against a pre-stated outcome. Record implementation quality, sample, uncertainty, guardrails and practical effect—not only a winner label.
Qualitative evidence
Customer interviews, usability sessions, sales feedback and support themes can explain comprehension, motivation and friction. Preserve who was studied, how evidence was collected and whether themes were consistent.
Operational decision
The change was judged on cost, capacity, compliance, brand or strategic fit rather than a performance test. Make the constraint explicit so a future team knows when the decision should be revisited.
Outcome asks “what direction did the result take?” Evidence asks “how do we know?” Confidence asks “how strongly should this guide the next decision?”
Clear answers
Frequently asked questions
Is qualitative evidence weaker than quantitative evidence?
It answers different questions. Qualitative work can provide strong evidence about why people understand or struggle, while quantitative methods estimate frequency or effect. Match the method to the decision.
Can a result have more than one evidence type?
Yes. A controlled result may be supported by interviews or sales feedback. Record the primary method and link supporting evidence without collapsing them into one certainty label.
Who chooses the confidence level?
The reviewer who interprets the result should state confidence and its basis. Teams should use shared criteria and allow later evidence to revise the conclusion.
Make this part of the next decision.
Diffined keeps the change, reason, evidence and conclusion together in one private history.
Try the working prototype ↗