OTTO and Exasol

From online retailer to platform – Exasol supporting otto.de on the road to becoming a data-driven company

OTTO is Germany’s largest online shop for home and living. Going back an impressive 25 years, the company was one of the first retailers in Germany to go online with otto.de with the goal of converting traditional mail-order business into digital business. Data plays a significant role in this process as insight into the needs and preferences of customers is used to generate tailor-made offers. As an infrastructure basis, an Exasol database guarantees performance at the required level when processing the ever-growing volume of data.


Challenges

Rapid growth of volume of data due to increase in products and customers

Solutions

Exasol was spearheaded as the data warehouse and high-performance database for the reporting tool Tableau

Benefits

Less than half the cost – savings coupled with better performance

Exasol’s analytics database proved impressive. Primarily due to an excellent price-performance ratio: low costs coupled with exceptional performance and integration with Loading...Hadoop.

Andreas Bonet Product Owner at OTTO
0 MIL.

Products

0 %

Lower costs alongside improved performance

> 0

Users can access the data at the same time

0 TB

of data

Background

One of the challenges for OTTO was converting traditional mail-order business into electronic business. Today, customers of the Hamburg-based e-commerce company can have their pick of more than three million products from over 6,800 brands at otto.de. More than 55 percent of purchases in the OTTO shop are made via smartphones and tablets. A factor of this success? Data handling. Even in the days of mail-order business, being savvy about the needs of customers was key. In today’s digital age, data analytics and business intelligence are crucial components of the corporate strategy.

  • Industry Retail and e-commerce
  • Application BI acceleration, Loading...Big Data analytics
  • Database Exasol
  • BI Tool Tableau
Case study background details

The journey

You can use our analytics database with our private cloud, any public cloud, on premises – or as part of a hybrid deployment.

The challenge

Dealing with large volumes of data

“Our customers are totally at home online,” says Andreas Bonet, Product Owner at OTTO. “They make purchases in a way that is more and more in line with their acute needs and use a variety of digital channels to do so. For us this translates to processing all the information the customer provides us with as part of data analyses in order to tailor personalised offers.” This is known as object-centric processing and it consistently puts the customer at the centre. The data from different channels is collated to form a comprehensive picture of customers in order to offer them an ideal shopping experience. Data protection plays the main role here. “This maxim is the key focus of every technical solution that we implement,” says Andreas Bonet.

Time and again, the sheer volume of data is presenting challenges for the data experts at OTTO. The growing number of products and customers is causing data points to soar. The need for storage space and maintenance grew, as did the desire for an answer: “We needed a solution behind our data warehouse which would provide our reporting tool with data in a stable yet speedy way,” says Andreas Bonet, summarising the basic requirement.

The solution

Proof of concept opened up new dimensions

The Nuremberg-based analysis experts at Exasol were tasked with implementing a proof of concept (PoC) with the aim of finding a solution. The first concern was to speed up data analyses and the process of drawing up reports on the basis of the available infrastructure. Exasol performed exceptionally well, particularly with regard to processing parallel requests. The PoC brought another issue to the forefront. “In the architecture planned up to this point, Exasol, as the access layer, would be provided with data from a data lake and a data warehouse to process it for reports. This called for the harmonisation of the data structures of both feeder systems which proved challenging,” explains Andreas Bonet. “Retaining twice as much data required too many resources and the loading path was too complex. We wanted to make positive changes here and wondered whether Exasol could function not only as an access layer, but also as a data warehouse.”

This brand-new architectural approach was put to the test in a second PoC and Exasol’s analytics database proved impressive. “Primarily due to an excellent price-performance ratio: low costs coupled with exceptional performance and integration with Loading...Hadoop,” says Andreas Bonet.

The Exasol database performs extremely well thanks to Loading...in-memory technology which involves data that is required regularly being persistently stored and processed directly in the main memory. This makes it predestined for analytical functions and large volumes of data. Thousands of users can access the data at the same time, as and when they need to, without jeopardising request throughput. Intelligent algorithms monitor the performance of the database and adapt the settings as required – significantly minimising administration outlay. What’s more, the Exasol database is designed in such a way that the broadest variety of programming languages can be integrated with ease via user defined functions (UDF).

Exasol as access layer and information stack

In the months that followed, Exasol was spearheaded as the data warehouse and high-performance database for the reporting tool Tableau. Loading...Hadoop was retained as a raw data store since the technology was ideal for Loading...data science activities and was heavily used by the developers.

“The project didn‘t just involve supplying our database,” recalls Mathias Brink, Customer Success Manager at Exasol. “OTTO had security requirements, for example, and we developed solutions for these together. We also implemented a spark connector as part of a jointly organised hackathon with OTTO, Exasol and partners.” “We have always felt that we are taken very seriously as a customer and our influence on Exasol’s product map is proof of this,” adds Andreas Bonet.

The benefits for OTTO

  • High-end performance thanks to Loading...in-memory technology
  • Quick analysis of large data volumes
  • Comparatively cheaper with a better performance
  • Scalable data volume
  • Easy and quick installation, intuitive handling
  • Integration with Loading...Hadoop, Tableau and various frameworks for data preparation
  • Integration of different programming languages via user defined functions
  • Cloud-ready

The future

Even more scalability required

Now that the infrastructure is designed for this purpose, Andreas Bonet and his team are keen to get the specialist departments even more involved. The ultimate aim is for them to be able to extract information from the data warehouse and analyse it according to their needs. In addition to the requested reports, the specialist departments will in future be provided with GUIs that they can use to access the data themselves. Only when each party knows what the other is doing can campaigns be coordinated – ensuring an optimal customer experience.

OTTO’s transformation into a platform opens up the company to even more brands and partners that sell their products at otto.de. In order to be able to process the increasing data volume smoothly in the future, the scalability of the solution, without soaring costs, was an important consideration right from the outset. Exasol’s solution is suited to all types of infrastructure, from on-premises operations through to hybrid and public cloud systems.

Start your Journey

Get in touch today

We’d love to hear from you if you’ve got a burning question for one of our experts – or if you simply want to talk through how to take your project forward with us, or one of our partners.

Contact Us