Authors: Felix Diekmann, Tobias Lenschen
As a manufacturer who sells products through dealers, you are no doubt already familiar with the topic of classification. This development is inevitable: traders demand information about products (= product data) that increasingly corresponds to a standard classification.
This request often encounters an infrastructure at the manufacturer's end that does not readily allow for the efficient handling of these requests. It is not without reason that there are various tools to, for example, re-qualify exported content from a product information management system (PIM) according to classification rules.
The problem is that such work is usually extensive and not very sustainable, since assortments and content change. However, this also means that classification versions and the associated rules change.
In short, the manufacturer is trying to meet the retailers' requirements without asking how they themselves could benefit from classifications. Might it even be possible to implement retail requirements more efficiently and flexibly in the future through classifications?
In this article, you can read:
What exactly does classification mean?
The classification of product data involves analysing and sorting the data into uniform groups or categories. Classifications are usually technically and professionally oriented and provide a clearly defined and unambiguous structure for the classification of products. The aim is to group similar products together so that rules or additional information can be applied to them.
An illustrative example of classifications in retail are the so-called product groups.
Product groups organise a retailer's entire product range into a clear hierarchy that groups products by category, such as type, purpose or material.
Take a hardware store, for example. Here, the items are usually organised according to a three-level classification:
- Super category (e.g. tools)
1.2 Subcategory (e.g. hand tools)
1.2.1 Spezifische Produktgruppe (z. B. Schraubendreher)
A screwdriver of a certain model could then be classified as ‘Tools → Hand tools → Screwdrivers’. This structure makes it easier for customers to quickly find the right product. At the same time, retailers benefit from being able to manage stock levels, reorders and assortment analyses more efficiently.
The following example* is based on the approach of collecting all screwdriving tools in the respective folders and thus describing them in a similar way:
Such classifications are not only crucial for store personnel, but also for digital logistics, especially in e-commerce. Here, a more refined classification is often used to offer detailed filter options, such as ‘screwdriver → cross recess → size PZ2’. This allows customers to search for specific features and find the desired product faster.
If you follow this approach when building a product data model in a product information management (PIM) system for your own portfolio, you will already be creating compatibility with standard classifications. This is also an important tool for ensuring that your own data quality is of a high standard.
What are the benefits?
more consistency in the content of product data through standardisation of descriptions (including consistent terminology and structure).
Simplification of data distribution to the various recipients, as retailers can more easily prepare and transfer data for marketplaces, for example. Markets can be more easily accessed or made accessible. Competitiveness and market presence increase.
easier product discovery and comparison for customers and business partners – even across different customer touchpoints
Increase in the efficiency and predictability of internal data maintenance through clear, measurable maintenance specifications that enable the targeted use of resources
more efficient realisation of translations
Support for the organisation of the range and selective data transfer; the groups formed offer filter options to optimise the procurement process.
What is a standard classification and what are the relevant systems?
Standard classifications aim to simplify the exchange of data between different market participants such as manufacturers, dealers, marketplaces and online shops. They can be motivated by different factors, e.g. by the industrial customer to optimise procurement, or by the manufacturer and wholesaler to be able to supply the sales channels of the trade better and more efficiently. Below is a list of the most relevant standard classifications:
- ECL@SS is a cross-industry classification system. It was originally developed by major companies such as Siemens, Bosch and Lufthansa to make C-parts procurement more efficient and to promote a technically supported central purchasing system. In this system, customers practically dictate to the supplier/retailer how products are to be classified and transmitted.
- ETIM was originally developed for electrical engineering. After the adoption of proficl@ss in 2019, ETIM also covers the areas of heating, ventilation and air conditioning (HVAC), plumbing, building materials, shipbuilding, tools and construction supplies. In contrast to ECL@SS, the classification is driven by manufacturers and wholesalers. There are translation bridges to ECL@SS versions that reduce the maintenance effort for manufacturers. For some time now, ETIM has also been working on the integration of a BIM standard (Building Information Modelling).
- GPC (Global Product Classification) was developed by GS1 and provides a uniform structure for the description of products to facilitate global trade. With the Global Data Synchronisation Network (GDSN), GS1 also operates the largest platform in which products must be classified according to GPC or ETIM. This classification has a high relevance and penetration in the consumer goods sector.
- UNSPSC (United Nations Standard Products and Services Code) was originally launched under the auspices of the United Nations Development Program (UNDP) and is now also managed by GS1. It is a global classification system that provides a standardised structure for describing almost all products and services. It is frequently used in international supply chains to ensure standardised purchasing and delivery chains.
In addition to these (most common) standard classifications, there are numerous others on the market. However, the same applies to all of them: a well-thought-out data model not only serves to organise and structure information, but also creates a basis for digital innovations, automation and an improved customer experience.
How can companies position themselves strategically in terms of the product data model?
For specialised data models, it is recommended to choose a technically oriented approach. This makes it considerably easier to transfer the model to the current standard classification. In certain situations, it makes sense to use a standard classification as a specialised data model.
In most cases, the standard classification does not cover all of a company's needs and a customised ‘house classification’ is a more sensible approach. The principle here is to be more granular than the standard classification so that a transfer is easier to carry out. This is also the basis for responding to future versions of the standard classification.
But what do classifications have to do with BIM?
BIM (Building Information Modelling) is the working methodology of ‘digital construction’. It enables a holistic view of construction projects over their entire life cycle when planning, constructing and operating buildings. This means that a construction project, including interior fittings, is planned on the digital drawing board, which results in an enormous increase in efficiency in the subsequent implementation. The basis for this is 3D models of the installed components – from sockets and pipes to boilers and ventilation systems, etc.
For manufacturers with a product portfolio that includes many variants, providing the information for the model can be a mammoth task, since a CAD drawing is not usually available for every possible variant.
The missing models can be created on the basis of technical features using an AI. This is where classification, which offers the perfect environment, helps to define the necessary features.
Standard classifications such as ETIM have also recognised the close relationship between the topics of BIM and classification and are already working on appropriate solutions (ETIM-BIM).
The use of classifications should be more of an end in itself for industry, as it also generates great added value in their own product data management. Thus, classifications no longer have to be perceived only as retail requirements. If companies focus on simply ‘meeting’ retail requirements, they deprive themselves of the many advantages that an integrated approach from the manufacturer's point of view brings.