Taste Evaluation

The Taste Problem

Design taste cannot remain tacit. In the age of AI-assisted product development, taste must be encoded as explicit evaluation criteria — operable, auditable, reproducible. Gut feel is the source from which taste criteria are derived; the weighted matrix is the mechanism that makes taste operational.

Key insight: Taste is a fourth lens applied after DVF, not instead of it. An option can be Desirable, Feasible, and Viable — and still fail taste by adding three new concepts to the user's mental model when zero would suffice.

Phase 1: DVF Filter — Primary Triage

Apply IDEO's three-lens filter first. Any option failing two or more lenses is eliminated before taste scoring.

Elimination threshold: DVF total < 6 → option eliminated before taste scoring.

Phase 2: Taste Criteria — Four Apple/Jobs Principles

Apply these four criteria to all options that pass DVF. Each is scored 1-5 with explicit rubrics — no subjective override.

Criterion T1: Subtraction

"Innovation is saying no to a thousand things." — Jobs, 1997

Test: Could this option achieve its goal with one fewer feature/concept/step?

Criterion T2: Concept Count

"Simplicity is the ultimate sophistication." Cognitive load is a design flaw, not a user problem to solve.

Test: How many new mental concepts does a first-time user need to learn?

Criterion T3: Progressive Disclosure

Complexity must be staged proportionally to user readiness. Front-loading is a design failure.

Test: Does the first interaction expose only what's needed for the first use case?

Criterion T4: Speed-as-Trust

Perceived responsiveness is the primary signal users use to assess product quality and reliability. 75% of users who experience slowness do not return (Akamai).

Test: Does this option introduce latency, friction, or steps that erode the sense of speed?

Phase 3: Weighted Scoring Matrix

Assemble all scores into a weighted matrix.

Default weights (adjust per product type):

Final score = Σ(criterion score × weight). Max = 5.0.

Output table:

| Option | DVF | T1 Sub | T2 Concept | T3 Prog | T4 Speed | Weighted Total |
|--------|-----|--------|------------|---------|----------|----------------|
| A      | 4.0 | 5      | 4          | 3       | 4        | 4.05           |
| B      | 3.3 | 3      | 5          | 4       | 5        | 3.84           |
| C      | 4.7 | 2      | 3          | 3       | 2        | 3.28           |

Phase 4: Recommendation

Produce top 3 options from the scoring matrix.

For each of the top 3, provide:

### Option [Name] — Score [X.XX]

**Why it scores well**: What taste principles it satisfies strongly
**Core trade-off**: What it sacrifices (every option trades something)
**Key risk**: The assumption that must be true for this to work
**Hire criteria**: Under what circumstances would a user choose this?

Recommendation: Identify the top option with a one-paragraph rationale grounded in the scoring — not preference. If the top option has a critical weakness, flag it explicitly.

Anti-Patterns in Taste Evaluation

DIVERGE Output for Taste Phase

Produce docs/feature/{feature-id}/diverge/taste-evaluation.md and recommendation.md:

taste-evaluation.md:

1. DVF filter table (eliminations documented)
2. Weights selected and rationale
3. Full scoring matrix for surviving options
4. Score breakdown per criterion per option

recommendation.md:

1. Top 3 options with pro/con/risk/hire-criteria
2. Recommended option with rationale
3. Dissenting case (which option the scoring almost chose instead, and why)
4. Decision for DISCUSS wave: "Proceed with [option], assuming [key risk] is acceptable"

Gate: Recommendation must be derivable from the scoring matrix. Any mismatch between scores and recommendation must be explicitly justified with weight adjustment.