Concept card

Graphical vs SQL vs Script Views

As of 2026-07-24T14:00:00Z

Edited by Cédric-Olivier Mâry, sole editor and publication director — 27 years in the SAP ecosystem. How these pages are produced, and where automation is used.

What is Graphical vs SQL vs Script Views?

Graphical views silently break HANA push-down past 5-7 joins — the single most common cause of dashboards that slow down six months after go-live.

What it is

Datasphere offers three ways to build a view over data: Graphical Views, built visually by dragging and joining tables on a canvas; SQL Views, written directly in ANSI SQL with SAP HANA extensions; and Script Views, written in procedural SQLScript for logic that plain SQL cannot express. All three produce a queryable object inside Datasphere, but they get there through fundamentally different execution paths, and that difference is the whole story.

Why it matters

The choice between the three view types is one of the few Datasphere decisions that directly determines runtime performance months after the modeling work is done, because one of the three — Script Views — does not push down to HANA Cloud's native execution engine the way the other two do. A modeling choice that looked purely cosmetic during a workshop can become the root cause of a dashboard that silently degrades as data volume grows, discovered only when users start complaining.

How it works

Why it matters in practice

Graphical modeling turns into unreadable spaghetti past 5-7 joins with no windowed-analytics support — fine for prototypes, wrong for production.
SQL views cover 80% of production-grade modeling; the default should flip to SQL once join count exceeds 5 or a window function is needed.
Script views are the procedural escape hatch — reach for them only when Graphical and SQL genuinely can't express the logic.

Key points

Three view types: Graphical (low-code), SQL (workhorse, default), Script (escape hatch — penalises push-down).
Default to SQL once joins > 5 or window functions appear. Graphical is for prototypes + simple dim wrappers.
Script view in Analytic Model lineage = 10-50× slowdown for every consumer query.
Three-layer composition pattern: staging (typing + currency) → domain (joined + named) → consumption (pre-aggregated).
HANA SQL surface includes full ANSI window functions, recursive CTEs, PIVOT, LATERAL joins.
Hierarchy artefact in the Analytic Model beats recursive CTE — let HANA evaluate natively.
Joule-in-Datasphere (H2 2026 GA target) will narrow the SQL skill barrier by drafting from natural language.
2 SQL-fluent modelers ship 3-4× the production content of 4 Graphical-only modelers after 6 months.
Graphical vs SQL vs Script Views is mastered only when it changes a named buyer decision.
Start with the semantic contract and control model before demonstrating the tool.

Terms used on this page

Graphical View: Drag-and-drop canvas-based modeling surface in Datasphere. Translates to SQL at runtime.
SQL View: ANSI SQL + HANA extensions, written as code. Pushes directly to HANA optimizer. The workhorse.
Script View: SAP HANA SQLScript with procedural blocks. Forces row-by-row evaluation, breaks push-down. Edge cases only.
Push-down: Query execution pattern where Datasphere translates SQL into HANA-native operations executed in column-store engine.
CTE: Common Table Expression — WITH clause that defines a named result set for use in a single query.
Window function: Aggregation that operates over a partition of rows defined by PARTITION BY + ORDER BY (e.g. ROW_NUMBER, SUM OVER).
Three-layer composition: Canonical Datasphere pattern: staging → domain → consumption views, each reusable and testable.
Calculation View: Legacy HANA-native view artefact (predates Datasphere). Datasphere SQL views compile into calc views under the hood.

Sources

Full card available to members. What the full card adds: the full decision framework · the SAP vs Snowflake / Databricks / Fabric comparison · the common pitfalls and their fix · the cheat sheet · the architecture schemas · the code blocks · the facts worth quoting.

Open in the app →