filter attributes fill rate
products per batch onboarded
- Young marketplace with great growth ambitions.
- Strong market interest in the platform.
- Broad product line with high data quality as target.
- Supplier and product data very heterogeneous.
- High manual effort for product categorization.
HOW ONEDOT HELPED
- More efficient onboarding of new suppliers.
- Automated product data categorization.
- Product data refined and enriched.
- Improved data quality and more products faster in the shop.
In April 2017, Österreichische Post AG launched the online marketplace shöpping.at with the aim of fostering the Austrian economy. The offering comes from Austrian retailers and manufacturers and ranges from small speciality producers to traditional retail chains. The most popular products at the moment are clothing, electronics, indulgence, home and garden as well as books. Online purchases are delivered via short transport routes exclusively by Österreichische Post AG, sustainable and CO2-neutral.
The Austrian online marketplace offers small and large Austrian retailers access to the digital market in order to sensitise consumers to local brands and suppliers. Extensive and detailed product data are required for the stationary and online trade to be successful on the Austrian digital platform.
The online offering is to be expanded rapidly and the onboarding of dealers and manufacturers is to be speeded up in order to be able to offer more articles on the Austrian online market. The product data of the suppliers are to be automatically integrated into the target data structure of shöpping.at and the quality of the product data is to be increased. These steps should accelerate the growth of shöpping.at and increase the conversion rate.
Process automation was a central success factor for the further increase in efficiency of shöpping.at. This is only possible with state-of-the-art software in this area. That is why the cooperation with Onedot makes sense for the Austrian online marketplace. Another reason for the cooperation is the possibility to use artificial intelligence and machine learning for data processing via Onedot in order to expand the desired process automation with this promising technology.
“Through Onedot and its AI technology shöpping.at was able to accelerate the supplier onboarding of large retailers enormously and thus help to expand the market growth.”
Andreas Pinterits, Head of Product, Strategy and Onboarding at shöpping.at
shöpping.at enables suppliers to make the extensive product catalogs available in their own formats. Onedot has read in these different formats, corrupt data records have been identified and sorted out. Since the supplier data was partially not very structured, attributes had to be extracted from the product texts first. The product data was then assigned to the online categories on shöpping.at. The system thus makes it easier for the online user to find the products. The shöpping.at data team reviewed the Onedot AI’s suggestions and overridden incorrect assignments. This feedback helps the software to learn and to perform such categorizations more accurately in the future.
The available and extracted attributes were automatically mapped to the target attributes of shöpping.at. Since each supplier designates its own attributes differently, one mapping per supplier had to be carried out. To ensure that search filters can be displayed uniformly and consistently in online shops, value lists, numbers and units were standardized. In order to present the various different configurations of the product in a user-friendly manner, product variants were created on the basis of the relevant attributes, such as size or color.
Suppliers are onboarded in individual batches. A batch depends on how many new products the suppliers have provided at that time. A representative batch contains approximately 100,000 products or more for the marketplace setup. In the four-level categorization, an accuracy of 94% was achieved in two iterations. The automatic mapping of the Onedot software offers the possibility to transfer more attributes compared to the manual mapping. These attributes can then be used for additional filters and further simplify the product search. On average, suppliers provided only 26% of the filter attributes in a structured manner. By refining the data, the Onedot software was able to increase the fill rate of the filter attributes from 26% to 86%. 97% of the products could be combined in possible product variants. This high-quality product data provides the basis for a successful customer experience on the marketplace, better conversion and more sales.