How Target2Sell Offers Ultimate Personalization Customer Experience With DataStax EnterpriseJune 13, 2017
This post is one in a series of quick-hit interviews with companies using DataStax Enterprise (DSE) for key parts of their business. In this interview, we talked with Guillaume Drot, who leads the the engineering team at Target2Sell.
DataStax: Hello Guillaume, thanks a lot for your time today. Can you tell us about Target2Sell and your role with the company?
Target2Sell: Target2Sell is the European leader in real-time one-to-one predictive marketing solutions for ultimate personalization customer experience. Our SaaS platform allows our merchants to offer the most relevant products to each of their visitors in real-time and on all digital channels. It uses latest machine learning techniques, deals with huge volume of data and computer analysis, and provides a real-time service to our merchants.
Target2Sell is used in more than 15 European countries and currently has over 150 customers, including Auchan, Sephora, Micromania, But, Camaieu, Catimini, Princesse Tam Tam, Raja, Manutan, Lapeyre, Ticketea, and Phonehouse.
I am leading the engineering team that includes back and front developers, and data scientists.
DataStax: What differentiates Target2Sell from similar vendors that offer predictive marketing software?
Target2Sell: Technically, Target2Sell first embeds over 900 algorithms on our servers coming, results of the latest researches from the most advanced mathematics universities of the world. Secondly, in order to be able to put in motion all those algorithms, we are able to process both transactional merchants’ data (sales data) as well as navigation data – both cold (past) and hot (in real-time). Third, we built an infrastructure to deliver personalized product merchandising proposal to each and every customer in less than 50 milliseconds. And forth, our machine learning enables to seamlessly evaluate best performing algorithms all along the run of the solution. The result is a ultra-personalized service that is the most performing in Europe to boost our customer’s’ revenues – we won all our A/B tests again in 2016 on performance levels.
DataStax: Did you use a different technology before you started using DataStax Enterprise (DSE)?
Target2Sell: We used and we still use MongoDB but it showed limitations when it comes to scale massive data inserts. We looked for a scalable database that can handle our insert/update throughput. Cassandra, and then DSE was a natural choice but MongoDB is still used for other use cases.
DataStax: Why did you decide to use DataStax Enterprise? What kind of data is stored in DSE?
Target2Sell: DSE allows to scale very easily by adding new servers. Datacenter replication allow to specialize servers for different usages and use cases. Today, we do not fear to reach a limit with DSE because we know we can add servers when necessary. Furthermore, DSE cluster is always online and thus ease hardware and software updates without adding any downtime.
We store our users information and tracking data in DSE. It is the data source for our learning algorithm that allow us to recommend the right product at the right time.
DataStax: How would you sum up the benefits you’ve achieved with DataStax Enterprise?
Target2Sell: DSE is highly scalable and reliable. It allows us to grow easily and helped us to build a zero downtime service.
DataStax: What caused you to use DSE over open source Apache Cassandra™?
Target2Sell: First, we wanted some support on the product. The DSE startup program is really helpful and gave us very good insights on how to implement our use cases. Then, DSE is shipped with Ops Center which is a very powerful monitoring and management tool and it freed us to develop our own tools.
Furthermore DSE ensure that the Cassandra version shipped is battle tested and provide enterprise grade support.
DataStax: What features from the DataStax Enterprise stack are you using at the moment? What business / customer experience outcomes have you achieved by using DataStax Enterprise?
Target2Sell: We mainly use Cassandra core features, but we started to use Spark more than a year ago. It was very easy to setup and it helped us to build more complex algorithm we can now provide to our customer.
DataStax: Tell us about the future of your project(s), do you intend to leverage other parts of DSE to make it a reality?
Target2Sell: Our main projects are to improve our usage and knowledge of Spark and to upgrade our cluster to DSE 5 in order to benefit of the new features and performance improvements.
DataStax: What advice would you give to other startups that are thinking about using Apache Cassandra™ and DSE for the first time in their solutions?
Target2Sell: I think startups should understand in details how Cassandra and DSE works in order to avoid long term problems. Data modeling can be tricky, capabilities of Cassandra are awesome but complex. They can rely on the DataStax academy and on the startup program to help them build performant and resilient software.
SHARE THIS PAGE