Data Migration from PIM Systems Made Easy

Tips for Keeping Parallel Operation as Short as Possible

When migrating a PIM system, it is necessary to replace the PIM system itself, as well as all existing output channels, interfaces, and functions, in a timely manner to keep the (time-consuming) parallel operation as short as possible. I would like to share from practical experience which approach has proven effective in this regard.

2.16.1.2

First and foremost, it is important to determine whether target systems - such as websites, online catalogs, or shops - will continue to be used in the coming years, or whether a transition, relaunch, or similar change is already planned.

• Are there any other relevant target systems that could not be integrated so far for technical reasons, such as configurators, classifications (ECLASS, ETIM), portals, or similar systems? If so, it naturally makes sense to include these in the planning as well.
• From a technical standpoint: Which interfaces and formats have been used so far? What “mappings” are necessary to ensure compatibility with existing integrations after the transition?

The following questions arise regarding the output of printed sales documents or PDFs that are already produced fully automatically using InDesign:

• How can the existing templates and placement logic for tables and pages continue to be used? Ideally, they already exist in InDesign.
• Where and to what extent does the content of the documents differ from the PIM data? Ideally, all content is already contained in the “old” PIM system and can therefore be easily migrated as well.

Tips for Data Migration

As already described in the section on data extraction, a step-by-step approach is not recommended. All data must be migrated “in one go” to keep the parallel operation mentioned above as short as possible. The advantage of this is that you have a complete overview of all the data, which makes the process more efficient and allows you to identify structural data issues more systematically.

From a technical standpoint, I would definitely recommend Excel for the data migration phase—for the simple reason that you can adjust, optimize, and supplement the data there. Incidentally, the ideal scenario is when the Excel migration files are based on a template (XML Schema XSD), which enables validations and conversions to XML.

However, before data migration can begin, you must verify whether the data can even be exported from the “old” PIM system in Excel format. Based on my practical experience, I can confirm that this has always been the case so far.

Using AI for Data Migration

Finally, I’d like to share my experiences with AI-assisted data migration. In my opinion, AI can reach its full potential only when the source data is already of high quality. Can AI also be used for data optimization? Theoretically, yes - the question is simply whether it’s cost-effective.

Data migration typically takes several months on average. So, AI might save you a few weeks. On the other hand, you have to keep in mind that a correction run is necessary, and corrections are often more time-consuming than manual processing.

Therefore, I would recommend performing data optimization primarily by hand and using AI in the context of data migration - for example, to check data quality and for simple tasks such as separating numerical units or extracting text modules.