Overview
A scale for grading how much to trust a visibility finding, from anecdote to controlled test.
Business problem
Recommendations mix hard data and hunches with no way to tell which is which, so confidence is misplaced.
Decision supported
How much weight to put on a given finding before acting.
Inputs & outputs
Inputs
- A claim or recommendation
- Its supporting data
Outputs
- Evidence level
- Confidence score
Step-by-step process
-
1
Locate
Place the finding on the ladder: anecdote, correlation, replicated pattern, or controlled test.
-
2
Label
Attach an explicit evidence level and confidence score.
-
3
Weight
Let low-evidence findings inform experiments, not big bets.
Maturity model
-
L1
Unlabelled
Findings stated without grading.
-
L2
Graded
Evidence level attached to claims.
-
L3
Weighted
Decisions scaled to evidence.
-
L4
Tested
Big bets require top-rung evidence.
KPIs
- % of claims graded
- Decision reversal rate
Common mistakes
- Treating correlation as proof
- Ignoring confidence intervals
- Betting big on anecdote
SearchScore insight
Recommended next steps
Where this fits - and what's next
The SearchScore path from a problem you feel to visibility you can measure.