Simon Edel
About the migration, handling and maintenance of data in PIM systems
The patient is virtual, characterized by meaningful attributes and neatly classified. Clear traces of use are evident. This is more or less how a PIM manager imagines a healthy data set. Surprisingly, the actual situation is often quite different. An independent team at Laudert has now been established to handle this type of case.
The overlooked patient
Why data in PIM systems is often inaccurate
The starting point is usually the introduction of a new PIM system. A number of data sets need to be migrated. The goal is a clean, unified and detailed data basis to develop a successful, high-performance omni-channel strategy for the future that considers shops, print media and all other channels.
When introducing new IT systems, there is one common mistaken diagnosis: “We already have good data that we have been using for years now.” But is the data actually good? After all, the goal of the new system is to create extended possibilities for application and clearer structures. That’s reason enough to want a more accurate clinical history.
It’s necessary to resolve not only which data needs to be migrated, but first and foremost how the future data model will be structured. This is defined based on the customer requirements and individual applications. Depending on the situation, complicated data assimilation or data enhancement may be necessary with respect to attributes, categories and classifications. In any case, a clean data model is a basic prerequisite for successful migration.
Classic error sources during migration include values with measurement units. Simple mix-ups, for instance between grams and kilograms, can result in disastrous consequences from incorrect packaging calculation to serious miscalculations in dealing with customers. In case of doubt, the customer will simply perceive this data as embarrassing because it is clearly incorrect. Not a great start for a new PIM.
Intensive care needed
Realistically estimate expenses for data maintenance and data
Realistic estimation of expenses is also fundamental for migration. Not all process steps can be automated. In particular, reading data from freehand texts, technical drawings or tables often requires a human eye. And when it comes to product volumes in the hundred thousands…
One preventive measure for data migration is a clean workflow that clearly defines human work and machine work. This process is always designed individually based on the requirements in the specific case. Such workflows describe not only the tasks to be performed or technical resources but also measures to preserve high quality of data.
Easy maintenance thanks to the content team
Laudert makes sure PIM data is always up to date
Although Laudert works intensively with (AI-based) automation of these processes and similar processes, the complexity of data migration, data maintenance, enhancement and creation still requires the work of trained specialists.
That’s why Laudert employs an exclusive content team that works together with workflow experts for high-quality data handling in system migration and for preparing existing data or adjusting inaccurate data sets. In addition, the team has expertise in the area of exporting and controlling data and acts as a liaison with media production, IT and all other parties involved.
The strategic objectives of a PIM system can only be achieved if the patient is healthy: Unified data that is maintained correctly, accessible at any time. This is the only way to enable efficient and successful communication both internally and externally without migrating “unhealthy” data or making a mistaken diagnosis.
