14. Plasticity

Definition

Plasticity is the ability of a data model to adapt to structural changes, new requirements, and unforeseen scenarios without losing coherence, stability, or analytical meaning.

A plastic model does not break when the business changes.
It reconfigures.

In Power BI, plasticity is manifested when the model can evolve (new dimensions, metrics, granularities, or rules) without requiring traumatic redesigns or massive rewrites.

Nature

Evolutive, structural, and strategic.

Plasticity is not improvisation.
It is anticipatory design.

A model can be:

• consistent,
• fluent,
• timely,

and still not be plastic if every change requires rebuilding it from scratch.

Function

To allow the model to:

• absorb business changes,
• incorporate new questions,
• and extend its useful life without degradation.

In Power BI, plasticity:

• reduces technical debt,
• protects analytical investment,
• and enables scaling without structural friction.

Consequence

When plasticity exists:

• changes are integrated at low cost,
• the model ages well,
• analytics evolves alongside the business.

When it does not exist:

• every change generates urgency,
• the model becomes fragile,
• analytics stagnates or is discarded.

The lack of plasticity does not fail immediately.
It fails over time.

Signals of Plasticity (✔️)

• New metrics are added without breaking existing ones.
• New dimensions do not force critical DAX rewrites.
• The model accepts new granularities.
• Business changes do not generate technical panic.
• The model “grows” without losing shape.

Signals of Rigidity (❌)

• Every change requires a full refactor.
• DAX tightly coupled to a fixed structure.
• “Special” tables for every scenario.
• Fear of touching the model.
• Phrases like: “that wasn’t planned”.

🔧 Samples — Plasticity in Power BI

🔹 Sample 1 — New dimension

❌ Without plasticity:
Adding a new dimension breaks existing measures.

✅ With plasticity:
Star schema + decoupled DAX → the new dimension integrates without impact.

👉 The model accepts growth.

🔹 Sample 2 — Granularity change

❌ Without plasticity:
The model only works at a monthly level.

✅ With plasticity:
The same model supports day, week, and month.

👉 Analysis adapts without duplicating logic.

🔹 Sample 3 — Business evolution

❌ Without plasticity:
Each new KPI requires parallel tables and measures.

✅ With plasticity:
Derived KPIs reuse existing structures.

👉 The model learns instead of fragmenting.

🔹 Sample 4 — Plasticity in DAX

❌ Without plasticity:
Rigid measures with hardcoded filters.

✅ With plasticity:
Modular, reusable, and parameterized measures.

👉 DAX stretches without breaking.

🔹 Sample 5 — Anti-pattern vs Pattern

❌ Anti-pattern — Rigid model

• Extreme coupling.
• Costly changes.
• Limited evolution.

✅ Pattern — Plastic model

• Structural decoupling.
• Incremental changes.
• Continuous evolution.

📌 Practical rule:
If every change requires redesigning the model,
the model is not plastic.

Interactions with Other Properties

• Consistency prevents semantic deformation during adaptation.
• Fluency sustains experience throughout change.
• Persistence protects historical continuity.
• Plasticity enables evolution without rupture.

Synthesis

Plasticity is not noticed when the model works today.
It is noticed when it continues working tomorrow.

A plastic model:

• does not resist change,
• does not break with evolution,
• transforms without losing identity.