ONLINE MARKETPLACE FOR LIFE SCIENCE PRODUCTS

The online biotech marketplace ZAGENO has exceeded its planned growth course for 2018, primarily due to its unique customer-oriented corporate culture. Committed ZAGENO expert teams from the fields of science, e-commerce, system integration and customer support work closely together and directly with customers. This enables ZAGENO to achieve smarter and faster processes, allowing more time and resources to be spent on value-added science and better research results.

The ZAGENO database contains 1,800 brands and more than 2 million stock units. This makes ZAGENO the leading online marketplace for life science and biotechnology products. High-quality product data and product attributes are crucial for researchers to find the right product that meets their experimental requirements. ZAGENO is also building the first comprehensive e-commerce platform for biotechnological products with cross-manufacturer product comparison.

To date there are no uniform product catalogues for Research Materials. Incoming supplier data and the present ZAGENO database have a variety of formats, attributes, structures and also different levels of data quality and data accuracy, which makes it difficult to integrate them.

HOW ONEDOT HELPED

ZAGENO uploads the raw product catalogs of new providers to the Onedot cloud. At a predefined interval, Onedot processes the uploaded product catalogs as a batch and provides ZAGENO with intermediate results for validation and feedback. One batch contains up to 600,000 SKUs.

The feedback from ZAGENO is collected and then used for further training and development of the Onedot models for machine learning. By repeating this process, the performance - both precision and recall - of the processing increases over time and the feedback effort required by ZAGENO is continuously reduced.

Each processing operation includes the following results:

• Matching of incoming products to ZAGENO's marketplace categories
• Assignment of supplier attributes to ZAGENO's target structure
• Extraction and normalization of key attributes such as sample size and units according to ZAGENO's data model
• Create a Golden Record from multiple data sources for product, variation or price level.

THE RESULT

Life science researchers will benefit significantly from the increase in data quality and the accuracy of categorisation. Through faster and more precise localization of products, ZAGENO will be able to further increase customer satisfaction while delivering the best possible search results. The automatic addition and categorization of up to 600,000 SKUs per batch will dramatically accelerate the onboarding process and go-live of the new major vendor.

With a precision and retrieval accuracy of over 95%, it is possible to successfully categorize incoming products into over 140 categories. In addition, only a few iterations and minimal training data are required to train the classification algorithms of Onedot. Golden Records are identified using Fuzzy Matching via predefined product attributes. Onedot also shortens time-to-market, improves conversion rates and reduces operating costs.

As an ambitious and fast-growing company with a focus on customer satisfaction, ZAGENO plans to expand its sophisticated biotechnology product portfolio to over 5 million products next year. Hundreds of new suppliers must be automatically connected and millions of incoming products categorised in order to cover ZAGENO's growing biotechnology product portfolio. This means that database management must also reach a new level of quality assurance. All these challenges are solved by the application of Onedot's data preparation service.