Concept card

SAC Smart Insights

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 SAC Smart Insights?

Smart Insights lets a business analyst self-serve 70-80% of typical analytical questions same-day instead of queuing 4-6 weeks for the data-science team — the genuinely hard 20-30% still belongs in Databricks.

SAC Smart Insights is the umbrella name SAP uses for the predictive and AI-augmented analytics features built directly into SAP Analytics Cloud: Smart Predict for time-series forecasting, classification, and regression; Smart Insights for automated chart commentary and driver analysis; Search-to-Insight for natural-language query against a story; and Smart Discovery for automated correlation and segmentation discovery. Together they form the business analyst's toolkit for applied statistics and machine learning, sitting inside the same tool an analyst already uses to build a dashboard, in contrast to a data-science platform such as Databricks, which is the data scientist's toolkit for building and operating custom models.

Why It Matters

Most analytical questions an organization asks are not exotic. What will next quarter's revenue be, given the last two years of history? Which customers are likely to churn based on the pattern already observed in customers who left? Why did this KPI move, and which dimension explains most of the movement? These are common, recurring, moderate-complexity questions, and historically they all had to queue behind the same central data-science team that is also asked to build genuinely hard, custom models. Smart Insights exists to let a business analyst answer the common questions directly, inside the BI tool, without opening a ticket and waiting weeks for a data scientist's attention.

How It Works

Why it matters in practice

Offloading forecasting, segmentation, and classification to citizen-data-scientist tooling frees the central data-science team from median-complexity backlog.
The trade-off line matters commercially — Smart Insights is what gets quoted in 70% of mid-market/Tier-2 conversations where Databricks-grade ML is overkill.
Search-to-Insight's natural-language query layer inside SAC Stories reduces the analyst's need to learn dashboard mechanics for common questions.

Key points

Four components: Smart Predict · Smart Insights · Search-to-Insight · Smart Discovery.
Citizen-data-scientist coverage: 70-80% of typical analytical questions.
Five canonical use cases: forecast · churn · pipeline regression · variance · NL self-serve.
Sample-size minimum Smart Predict: ≥ 24 monthly · ≥ 100 daily points.
Search-to-Insight quality depends on AM design discipline (companion C001/C018).
Escalate to Databricks ML for multi-feature · real-time · custom architectures.
Joule + Smart Insights surface ownership design needed (companion C024).
EEA-resident SAC processing for CNIL-aligned PII training.
SAC Smart Insights 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

SAC Smart Insights: SAC's predictive + AI-augmented analytics layer covering forecasting, commentary, NL queries, discovery.
Smart Predict: Time-series forecasting · classification · regression on customer datasets. Citizen-data-scientist grade.
Smart Insights commentary: Automated explanation of chart anomalies — driver analysis surfacing why a value moved.
Search-to-Insight: Natural-language query interface within SAC Stories. Returns answer + chart inline.
Smart Discovery: Automated correlation + segmentation surfacing patterns without explicit hypothesis.
Citizen-data-scientist: Business analyst empowered by tooling to deliver typical analytical questions without central data-science queue.
Sample-size minimum: Smart Predict reliability threshold: ≥ 24 monthly points or ≥ 100 daily points.
Databricks escalation: Multi-feature · real-time · custom-architecture cases routed to Databricks-in-BDC + MLflow.

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 →

Guides that answer with this page

These guides cite this page as one of the sources their answer rests on.

SAP Analytics Cloud Predictive: Scope It Honestly