Customer Story
Medion new
Discover how MEDION was able to increase query performance by an impressive 70% and save several hours of data processing overnight.
flaconi & Exasol
The Beauty of Data: Exasol is Supporting flaconi to Excel as a Data-Driven Business
Everything Related to Data Comes from Exasol
Discover how flaconi, in partnership with Exasol, was able to quickly adapt to changing business sectors and scale in line with rapid growth.
flaconi yielded significant benefits:
One of Germany‘s Leading Online Beauty Retailers
Location
Germany
Employees
450+
Industry
Retail & E-Commerce
Product
Exasol Espresso
flaconi, which is part of ProSiebenSat.1 Media SE, is one of Germany‘s leading online beauty retailers. Founded in 2011, the shop offers a comprehensive range of fragrances, skincare and hair care products, make-up, natural cosmetics and accessories. Customers can choose from over 720 brands and 45,000 products. More than 450 employees are at the forefront of this “unconditional customer service” and want the business to become the most innovative and customer-friendly online beauty store. The first important milestone on this journey was winning second place in the 2017 Online-Handels-Award trade awards. From the very first day of its existence, flaconi has adopted a data-driven strategy – both internally and in the way it approaches customers. The database used was set up to withstand the rapid growth of an ambitious start-up. However, the system had already reached its limit by the time the foundation phase was over and was not built for the professionalization process over the next few years. flaconi’s decision to opt for Exasol‘s in-memory analytics database has taken the business to a new level and it is now ideally equipped for the future.
The challenge: performance, agility and scalability
There is enormous pressure in e-commerce. The main objective of each player in the market is to adapt quickly to changing business sectors and to be able to scale in line with rapid growth. No other industry is so data-focused – every single decision is based on data.
flaconi therefore followed a clear approach in the search for a new database solution: “We wanted a powerful and scalable analytics database to serve as the basis for our business”, says Martin Nguyen, Director of Strategy, Data and Analytics. “We were very target-oriented in our search and looked at the providers’ references in particular.
Aside from requirements such as performance, flexibility, scalability or low maintenance, the human factor of our partner was very important to us,” says Martin Nguyen, summarizing the selection process.
There were three clearly defined objectives. Firstly, the future database had to support business steering i.e. delivering data in the right format at the right time. Reports, real-time alerts and simulation models inform those responsible about certain performance indicators. Countermeasures can then be taken at any time depending on positioning and requirements. Secondly, another priority was creating “data products”. The plan is to optimize the customer experience in the long term using products supported by data science or machine learning, such as the recommendation engine, as well as product forecasting and pricing models. Martin Nguyen‘s team also wanted to enable flaconi‘s employees to use self-service BI, so they could access existing data easily as required and use this as a basis for making informed decisions.
After a short test phase in the form of a three-day proof of concept, they decided on Exasol‘s analytics database. “We have been especially impressed lately by the speed of the database, the product roadmap, the functionalities via UDFs (User Defined Functions) and the excellent communication with those responsible at Exasol.”
More users, faster reports
It took six months from setting up the server to going live. “The migration was very easy. Being able to genuinely transfer all the logics from the old database to the new one was important to us. We didn’t want to go live with less than what we had before. Of course, our new system should be better than the old one,” explains Nguyen. “As we have already enjoyed good experiences with Amazon Web Services, we are now opting to run the full database application on the cloud and AWS too”. flaconi uses Python, Pentaho Data Integration and AWS services for data integration. The company migrated from Power BI to Tableau to visualise the data.
The Exasol database currently hosts around three terabytes of data. Around 20 users have direct access to Exasol, with approximately 100 users using the database via appropriate dashboards, cubes and reports. “The number of users is rising steadily. With our “Data University” we are making sure that employees can keep improving the way they handle data,” says Nguyen. “Exasol has become our key tool. Everything we do is based on the database.” Work has also improved in terms of speed and keeping data up to date: “We used to get our reports around midday – now we have already received them by around 5.30 a.m. Certain ETL processes used to take 45 minutes with the old system and the same process takes no more than 15 minutes now,”, adds Nguyen enthusiastically. These standard e-commerce reports are extremely important as they help to steer the business directly. They provide information on how many purchases have been made and how many buyers and potential buyers have visited the website.
“All this data can be broken down to each individual marketing channel and we can find out exactly how often products are sold and what margins have been achieved. And we can produce a forecast for every single product for the coming weeks. Thanks to Exasol, we are extremely agile. In addition, we spend hardly any time on database administration – this used to take a very long time,” summarizes Martin Nguyen.
Data is Being Kept in a High-Performance Database
flaconi customers can tell that the data is being kept in a high-performance database in a number of ways. For example, if a customer clicks on a product, the likelihood that it is actually available is much higher than before. This is because forecast processes run on the database, so if a product is purchased, it is also automatically delivered to the warehouse at the right time. “Exasol also plays an important role in product recommendations. We decide which product to promote to which customer on our website or in Google ads based on product data and customer behavior. We recently set up highly centralized core processes as part of our supply chain for the product and customer experience and for marketing automation based on data. Practically everything related to data now comes from Exasol,” adds Martin Nguyen.
He points out that so far they have found very little in Exasol that is not optimally displayed. Martin Nguyen‘s is currently trying to find out how well suited, in terms of usability, the system is for flaconi data scientists. The data scientists draw their own data from the database and aggregrate it to a certain extent.
However, they build and train their data models on dedicated servers. “When it comes to performance and practicality, we are now trying to make the best use of the infrastructure that Exasol offers. We want our data scientists to be able to work in a familiar environment. This is something we will be stepping up further in the next few months,” explains Nguyen.
Try the Exasol Analytics Engine
From AI to DWH, we’ve got you covered.
Customer Stories
Discover inspiring success stories
