Concept card

SAP Data Services (BODS)

As of 2026-09-27

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 SAP Data Services (BODS)?

The classical SAP ETL tool — batch dataflows, a thick Designer client and a Job Server — still running quietly under a large number of SAP estates. It is where enterprise data-quality logic was written before flows moved into Datasphere.

What it is

SAP Data Services, still universally called BODS after the BusinessObjects Data Services name it carried, is the classical enterprise ETL tool of the SAP estate: a thick Designer client, jobs made of dataflows, a Job Server that executes them on a schedule, and a repository that holds the metadata. Alongside it sits SAP Information Steward, the profiling and data-quality companion that made BODS more than a mover of rows.

Its relevance is that it is still there. Data Services rarely appears in a strategy deck, and it rarely disappears from a landscape either. It is the tool that loads the warehouse overnight, cleanses the customer master, matches and deduplicates addresses, and feeds a dozen downstream systems nobody remembers subscribing. In a migration inventory it is routinely discovered late, because the people who built it have often moved on and the jobs simply keep running.

The data-quality logic is the asset, and the trap. Movement is the easy part of a BODS estate: a dataflow that reads a table and writes another has an obvious expression in a Datasphere replication flow. What does not move easily is everything BODS carries around the movement — address cleansing with reference directories, match-and-consolidate rules, custom functions written in the Data Services scripting language, and the accumulated exception handling that makes a nightly load survive bad input. That logic encodes years of business decisions, and it is usually undocumented outside the jobs themselves.

Why it matters

It is the undocumented layer a migration finds late. The cleansing and match logic inside BODS jobs encodes years of business decisions that exist nowhere else.

Key points

The classical SAP ETL tool: Designer client, jobs made of dataflows, Job Server, metadata repository.
Still called BODS after its BusinessObjects Data Services era; SAP Information Steward is the profiling and data-quality companion.
Straight movement re-expresses cleanly as a Datasphere replication flow.
🔴 Address cleansing, match-and-consolidate rules and custom scripting functions are the part with no clean successor — and the part that is undocumented.
Estimate by the ratio of data-quality dataflows to movement dataflows, never by the total count.
The decision is per job: straight-movement dataflows re-express cleanly; cleansing, matching and standardisation logic is the real decision.
Re-expressing a data-quality rule means reproducing it exactly (unverifiable) or re-agreeing it (needs a business owner who may no longer exist).
Sort the inventory by whose clock a job is on — source-system-triggered versus independent — before quoting anything.

Terms used on this page

Dataflow: The unit of work in Data Services: sources, transforms and targets assembled in the Designer.
Job Server: The Data Services execution component that runs jobs on a schedule.
Information Steward: The SAP profiling and data-quality companion product to Data Services.
Match transform: The Data Services transform that identifies and consolidates duplicate records — typically the least portable logic in the estate.
Dataflow: The Data Services unit of work: a source, a chain of transforms and a target. Counting dataflows is not scoping; classifying them by what they contain is.
Data-quality transform: A cleansing, matching or address-standardisation step. It carries business-agreed rules and is the part of a BODS estate with no mechanical successor.

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 →