Combining behavioural evidence with human creative standards

An advisory review system that explains creative observations, shows evidence, and preserves expert override.

Creative intelligence

Define rubric → Collect context → Assess criteria → Show evidence → Expert override → Learn disagreement

01Define rubric
02Collect context
03Assess criteria
04Show evidence
05Expert override
06Learn disagreement
01 / Problem
Frame the work

The situation I was solving

Creative review can be subjective and difficult to learn from, but behavioural performance alone cannot explain quality because audience, placement, offer, and timing also shape results.

Read the underlying principle: Taste is becoming a business capability.
02 / Value
Define what changes

What becomes better

The concept creates a shared rubric, explainable AI observations, evidence, uncertainty, and retained expert overrides—making review more consistent without pretending taste is objective.

03 / Approach
Design the system

How I work through it

I separate observation from judgment, version the rubric, adapt standards by channel and goal, and treat repeated disagreement between model and expert as learning evidence rather than error to hide.

Go deeper: A decision system beats another dashboard.
04 / Insight
Carry the learning

What I carry forward

AI should make creative judgment easier to articulate, not replace the person accountable for the brand.

Explore all educational field notes →