Leave computers to price and get the best price at all times: Dynamic pricing is where many online and multichannel retailers expect better revenue or more business. Technologically, dynamic pricing is already feasible. Based on sales and sales figures, computer programs are already predicting how demand and product cycles will develop and are pricing them – for different times of the day, weekdays, seasons, regions or individual cities.
Dynamic pricing can also be geared to customer groups. In this case, regular customers are granted higher discounts than new customers. The same applies to customers who regularly produce high shopping carts or low returns. Dynamic pricing can also attract potential customers with short-term discounts: thank computers and, above all, data.
Without data analysis no sales channel
Dynamic Pricing is just one example of the many opportunities big data and real-time analytics already offer. Many retailers also rely on the playout of personalized offers on the Internet and want to interlock different sales channels for Click & Collect offers. But such strategies are unthinkable without the evaluation of data. For analytic tools, ratings, and delivery or delivery processes to work across channels, the master data on which they are based must be accurate, consistent, up-to-date, and unique.
This is not enough in the new digital economy – not just in retail. This is the result of the study “revival of master data” of the consulting Lünendonk, which was developed in cooperation with the e-commerce specialists of KPS and the technology companies Salt Solutions and Zetvisions. According to the study, a whopping 85 percent of companies report problems with their master data, with their collection and evaluation: they complain about duplicates in the data sets, but above all, uneven data structures. In view of the increasing digitization this is simply a catastrophic finding, especially as the trade in Germany is competing with corporations such as Google and Amazon. These in turn have their data under control and surprise again and again with new, customer-oriented offers.
As a result, this two-part series addresses the problem of “master data”: The first episode deals with the importance of master data and the need for a uniform, cross-departmental strategy for data management. The second episode identifies appropriate measures for maintaining existing data and opportunities for optimizing the collection of new data.
Without good data quality, there is no success
Management of master data is certainly not easy in trade. The amount and complexity of the data grows with the number of products, their variations, different prices, but also with frequent changes in the supplier’s assortment. Last but not least, the number of data in multichannel trading is also increasing rapidly due to the linking of different sales channels. According to the study, the quality of master data differs on the basis of four central features: 62 percent of the surveyed retailers rate the timeliness of their data rather than bad. Data that is out of date binds working hours. The dealer can not rely on them – consistency is lacking. 68 percent of respondents observe gaps in their datasets; there is no completeness. Finally, different names prevent uniform classification and assignment of the data. Businesses are therefore too careless with semantics and expression.
Within a company, several departments collect data, often with different tools and programs. If a trader uses several distribution channels, the customer data in the shop, in the shop system, but also in the service department and in the customer service are usually collected in different programs. Errors can creep in at all interfaces: if, for example, names are not recorded uniformly, duplicates are created in the data records and the customer history can not be comprehensively considered across all channels. If category managers also name properties and product advantages in the shop system with trend words, they can not be clearly assigned in the merchandise management system. In the shop so the offer,
The biggest obstacle to the collection of master data is the lack of strategies and goals, a lack of management awareness, and insufficient communication between different departments. This is evident in the interaction between IT and specialist departments. Because in most companies, IT selects corresponding analysis tools and integrates them into the existing software architecture.
Data management lives from internal voting
The evaluation of the data is again the responsibility of marketing, purchasing, customer service and other departments. Also at this interface, communication and common procedures are usually lacking. While about every fifth IT employee considers the data questions to be solved, according to the study, only 12 percent of specialists from other departments share this opinion.
However, digitization is increasing the requirements for master data. Because now machines and mathematical models take over the analysis and need uniform and consistent data sets. For years, companies have been investing in software solutions to optimize their data management. But in the integration they fail to adapt the technology to their structures, but above all to determine personal responsibilities and formulate goals for the management and maintenance of the master data. Without a strategy, however, the data sets from different departments, distribution channels and electronic programs can not be standardized, so IT oversees the needs of marketing, purchasing and category management and vice versa.
If big data and analytics are to drive business, the data management strategy becomes a leader. It starts with the fundamental consideration of what data is important or could be important to the company and its business. For the formulated goals, responsibilities should best be identified in cross-departmental teams. Thus, a transformation process can start in which not only data is collected uniformly and maintained according to clear criteria, but the departments involved can restructure and become more digital. Data management can therefore be the beginning of a comprehensive change process. The data strategy should also contain clear guidelines to protect personal customer and user data. Who gets access to what data when,
Planning and optimization of master data are difficult to separate. The affected departments assume the content definition and assignment of master data, and the IT department adjusts the programs and tools used. But in data management, the views of all stakeholders should be continuously integrated. Therefore, it is advisable to place the responsibility for data management in cross-departmental teams. Once the parties have clarified what master data is needed in the future, they can develop methods for consistent collection and storage as well as the criteria for maintaining existing data. With these preparations, the data should improve soon and noticeably and the company can recognize the levers faster.