Die Omni-Channel Data Challenge

Project costs (digits)
Running costs
(per month, digits)
Die Omni-Channel Data Challenge
Unlock sales potential
Plug and Play
Project costs (digits)
Running costs
(per month, digits)
Jonas Christiansen
The problem
INTERSPORT wants to use its own data to increase performance, but this data is only available sporadically and locally.
Digitalisation provides retailers with vast amounts of data from a wide variety of channels: whether information from online shops and marketplaces, B2B channels of the specialist dealers or at the point of sale in the stores, from merchandise management, marketing tools or CRM systems. The challenge here is to make this data more usable in order to be able to generate real added value from it, to always have a clear view of sales, margin and profit and to ensure a cross-channel 360° shopping experience for every customer.The most successful medium-sized group in sports retail has more than 900 INTERSPORT retailers and must approach digitisation together with these individual independent retailers.
This involves a variety of data from a wide range of systems, online and offline.
Often there is a lack of comprehensive platforms, technologies, resources or know-how in data analysis and 'intelligent data use'.
The solution
A cloud-based business intelligence solution enables INTERSPORT and its retailers to make the best possible use of their data and thus establish a future-proof, data-driven working culture.
Through targeted analyses, the company not only optimizes workflows, but also better identifies market opportunities, serves customers more specifically and tracks inventories more effectively. Transparent data analysis, centralized reporting and a data-driven culture support the company in repositioning itself and thus expanding its competitive advantage.

A strength of the solution is not only the visualization of the data, but also the solution of data problems. Furthermore, it provides a holistic approach and covers the broadest range of applications with technology and tools in one solution. From standard and custom integrations, a high-performance ETL line and a flexible analytics database with a built-in commerce data model to the tools needed for a data-driven work culture, everything is included. A further differentiation is achieved by the fact that the solution is ready for use virtually from day 1 by means of straightforward onboarding via standardized interfaces. Desired individualization and extensions can then be implemented parallel to productive use. Thus business users from all areas can work data-driven - without explicit IT knowledge. A technological unique selling point is also the best-practice data model: Through the implementation of numerous use cases from the retail sector, the software has a uniform definition of key figures. The unique, comprehensive commerce data model comprises more than 500 attributes and key figures and serves as the basis for all tools and uniform communication - a perfect fit for every retail organization, regardless of industry and business model. In order to meet specific customer requirements, additional interfaces, tools or features are developed together with the customer on the basis of use cases, which are then available to all users. In this way, the company regularly supports its customers with new features and lays the foundation for being able to compete in the market.

For over 6 years and with the extensive expertise from 150+ implemented projects in retail, eCommerce and Omni-Channel, the provider supports its customers in data-based decision making. Already various companies from the fashion industry, such as Philipp Plein, Betty Barclay or Scotch & Soda, high-turnover retailers like the ANWR Group, Depot and Decathlon, but also innovative fashion start-ups like SugarShape and Avocadostore rely on the "Plug & Play" BI solution and thus control their business in a data-driven way.

  • no omni-channel transparency
  • no 360° customer profile: there is no database with complete customer profiles from online and offline touchpoints
  • no usable analytics: lack of analytical expertise to use / prepare data from complex IT systems
  • lack of data-driven processes
  • Full Omni-Channel Transparency: Integration of online and offline data sources for a holistic business view as well as efficient analysis of company data in order to derive targeted recommendations for action.
  • 360° customer profile: By integrating data from all relevant sources, INTERSPORT and the retailers have complete customer profiles that can be automatically fed into third-party systems for personalised marketing campaigns.
  • Self-service data access: The clear design and easy-to-use user interface makes it easy for users to intuitively find their way around, create reports and analyses and respond to new requirements without the need for collaboration with the company's IT department.
  • Data-driven work culture: Across all central work areas, employees are enabled to extract action-relevant information, make better and faster daily decisions based on data, and perfect processes using data.
Stumbling blocks
Even though users can access all relevant information independently in the "self-service" mode, questions remain unanswered from time to time - with a data model that includes over 500 metrics and key figures. In this case, the customer support team is there to help you with advice and practical assistance: whether it's requirements analysis, use case definition or hands-on support.

Support for employee acceptance

Use cases and user needs must be the focus of attention when introducing new systems in the company. It is therefore important that projects follow an interdisciplinary approach and that representatives from all user areas are involved in the overall process. The communication of goals and added value plays a decisive role in this.


Average KPIs for projects of the provider

The return rate can be reduced by 2-5%.

  • Reasons for returns can be identified and eliminated immediately through proactive alerting.
  • Marketing campaigns with high return rates can be reduced in favor of better performing campaigns and keywords.
  • Collection purchasing can be consistently oriented towards low returners with product segmentation.
  • Known return sinners can be treated individually with the help of customer segmentation (e.g. no more allowing multiple sizes of the same item in the shopping cart, no more free returns, no purchase on account, etc.)
  • High-conversion areas of the shop can be filled with high-margin and low-return products (automated integration via data feeds)

Customer value and turnover 

The customer value (CLV) of existing customers is maximized and new customers with the highest possible CLV potential are acquired at the lowest possible cost. At a conservative estimate, an increase of 10-15% on the CTMV is possible.

  • Existing customer ratio and CLV can be increased by automated churn prevention with proactive alerts and automated churn prevention segments that can be firmly integrated into marketing chains.
  • Analysis of patterns in the acquisition campaigns and products of customers with the highest CLV potential and corresponding campaign control for current acquisition campaigns.
  • Customized newsletter approach with individualized product recommendations and discount levels, depending on individual preferences and overall history.
  • Higher customer satisfaction through operational excellence. Prevention of system cancellations due to lack of inventory through timely proactive alerting about imminent stock shortages or in case of disruptions in the shipment of goods.

Profit margins

At a level of contribution margin accounting with an improvement of 2-3%, 10% more contribution margin from sales can remain.

  • Minimization of price markdowns in collection sales. By monitoring the sales and price markdown rates on a daily basis and knowing the individual price sensitivity of each customer, the price can be reduced in each phase of the sale so little that the customer converts without giving away any profit margin.
  • Negotiations with suppliers based on objective data with clear margin targets. Not as before with a theoretical margin, but on the basis of hard facts from real business.
  • Collection optimization in purchasing by eliminating low-margin products
  • Optimization of the conversion-strong areas of the online shop with high-margin products (also customer-specific possible).
  • Contribution margin-optimized marketing control to minimize the CPO (cost per order) and thus maximize the contribution margin.
  • General marketing optimization due to 100% transparency of performance and contribution margins, as well as all costs of all channels in a central system.
The project schedule
Step by step:

a) First, a requirements analysis with concrete use cases is developed together with the customer and potential users are identified. b) The Business Intelligence solution is presented and concrete solution approaches are visually illustrated within a demo. The customer also has the opportunity to test the solution himself via a demo access. c) After signing the contract, a workshop with the stakeholder groups takes place at the customer's site in order to implement use cases with specific users. d) Each customer has a dedicated key account manager who, in addition to technical support, is also available to advise on optimization and other application areas.

Data quality:

Customers who make multiple purchases as guests on a website do not have a complete customer profile and can therefore only be identified by their e-mail address, for example. In the software, these purchases can be assigned, consolidated and merged into a customer profile based on the e-mail address. This leads to a better and above all complete data basis for further analysis.

Where is the data:

Access via the web frontend is https-encrypted. In order to offer state-of-the-art technology with high operational and cost efficiency, the data is hosted in a fully scalable cloud environment at Amazon Web Services in Ireland. As a result, the customer pays only a fraction of what it would cost to own a hosting environment and does not need to invest any resources in operation and maintenance.

Project maturity level
Used company-wide
Project duration
The environment is ready for operation within a few days, delivers first results and can therefore be quickly tested and optimized in a productive environment.
Project cost
Running costs
(per month, digits)
Involved employees
(Operating phase, FTE)
Involved employees
(Project phase, FTE)
Skills required by the customer
Since real added value from data can only be achieved if working with data becomes a natural part of the daily work processes of all users, a solution is needed that is not only easy to integrate, but is also designed to be intuitive and user-friendly. The clear design and easy-to-use user interface makes it easier for users to find their way around intuitively, to create individually configurable reports and analyses and to react to new requirements without the need for collaboration with the company's IT department.
The project schedule
Step by step 

a) First, the customer's existing technical setup is recorded, including for example the source systems. b) This is followed by the definition and setup of the interfaces. c) Then the data exports are set up by the customer d) The provider validates customer data and creates user accounts e) The Business Intelligence solution is ready for use.


The solution has standard interfaces to connect systems via a simple "plug and play" approach. In addition to source systems (pixi Descartes, Spryker, Magento, Newstore, Google Analytics, etc.), outgoing interfaces to third-party systems - especially to operative systems such as email marketing or visualization tools such as Tableau - can also be connected.

The ETL process:

In the Extract phase, the data is extracted from all relevant source systems (e.g. Web Tracking, Google Analytics, ERP, marketing tools, etc.) In the Transform Phase, these different data sets are then brought into a uniform format and cleaned up, then consolidated and in the Loading Phase integrated into Elastic Search and made available for analysis. The ETL process is built in Apache Spark, an analytics engine for big-data processing that ensures high performance and efficiency and, most importantly, full scalability.

Data quality:

To improve data quality, validation ensures that data from the various sources is robust, reliable and meaningful. A method for the continuous improvement of the quality takes place for example in the Transform Phase. In order to avoid duplicates, it is ensured that the uniquely assigned ID of each data entry is contained only once in the database. If the data entry occurs more than once, the information of the last entry is used.

Jonas Christiansen (Senior Business Development Manager)
"If a customer is looking for the fly in the ointment, we can't help, but when it comes to generating real added value from data, we are the leaders"
Jonas Christiansen
The information may of course vary in individual cases. Please contact the provider for an assessment of your project.