Custom code: The burdened legacy from the SAP ECC era

The results of the latest DSAG Investment Report show a clear picture: for 79 percent of the companies surveyed, the profitability of investments in SAP software is a strong or very strong criterion. At the same time, S/4 migrations remain the most important investment driver; 42 percent of companies currently see them as a key initiative. This is precisely where a well-known problem comes to light. Over many years, an extensive stock of customer-specific code has accumulated in almost all large ECC system landscapes, which originally served to quickly cover process requirements, create differentiation options or close functional gaps. What was a pragmatic decision at the time later developed into a cumbersome technical legacy. Depending on the degree of individualization, up to 80 percent of the entire code base in ECC environments today consists of individual extensions. In highly individualized landscapes, between 60 and 80 percent of this is no longer used. The result is increased system complexity, excessive costs and increasing technical risks.

TCO under pressure

The effects are clearly noticeable: maintenance, error corrections and recurring adjustments cause high lifecycle costs and drive up the TCO in the long term. Upgrades become costly and resource-intensive projects because outdated custom code requires extensive testing and integration measures. At the same time, dependency on a small number of key technical personnel increases. In addition, there is a lack of transparency about which objects are being used or are business-critical. Finally, custom code becomes a central stumbling block for the S/4 transformation, as many older extensions are no longer compatible with modern architectural principles, interfaces and data models. As part of their migrations, many companies are therefore deciding to transfer their legacy code to the new S/4 environment virtually unchanged. In view of tight budget and personnel resources and the end of standard maintenance for SAP ECC in 2027, this approach initially seems pragmatic. It often remains unchecked whether the old enhancements are still needed or whether their functionality would have long since been covered by the SAP standard. However, this decision harbors long-term risks. The old custom code is based on operating models from previous years and can place a disproportionate burden on modern IT infrastructures. In combination with the long-term cloud orientation of the SAP strategy and the requirement for a clean core, this creates an environment that slows down the ability to innovate and jeopardizes release capability.

AI-based custom code analyses

This is where the latest developments in the field of artificial intelligence offer a promising approach. Modern AI-based tools, such as the solution from Nova Intelligence, now make it possible to automate large parts of custom code analysis and transformation. Agent-based processes can identify unused extensions and deactivate them in a controlled manner. They can also automatically transfer actually required custom code to a clean-core-compliant target environment. Companies should consider this new technological option before burdening their newly designed S/4 environment with legacy code from the ECC era. Such agent-based solutions can not only speed up the clean-up process, but also ensure that future enhancements consistently follow clean-core principles. The result is a modern, maintainable and future-proof system landscape that can exploit the full potential of S/4 and the cloud.

To the partner entry: