Tuenti needed to introduce a faster data analytics platform.
Tuenti mobile services are loved by customers who needs a totally different seamless communication experience doing voice, chat, messaging indifferently from an integrated mobile and web platform through a single phone number.
Tuenti must continuously perform detailed analytics to understand user behavior and feedback on integrated social communications platform on web and mobile to keep the customer base loyal. It also wanted to accelerate the analysis of customer behavior data, to reduce customer churn and drive user improvements. For social media and telco companies, understanding user behavior is vital to success. Detailed information on how people move through the websites and apps, and where the bottlenecks or problems are, enable an enhanced user experience.
The company had been using MySQL and Apache Hadoop to store customer data; however, performing advanced analytics was proving time-consuming. It wanted to find a faster way to process data and become more agile in response.
“From a big data analytic perspective, we have two main areas of focus: telecoms operations, such as advanced mining of call logs, SMS and mobile data consumption. The second is customer experience: who is engaging most with our services and how our app is being used,” explains Jaime Medrano Navarro, Lead Engineer, Tuenti. “The Tuenti seamless experience is our competitive differentiator and we are constantly adding, dropping and finetuning features based on a data driven analysis of customers preferences and usage, but we needed deeper granularly and faster insights for our ad-hoc analysis.”
Previously, Tuenti had been extracting data from production databases and generating static reports, however, this did not deliver the level of insight required: “We performed batch reports, but they were summarized and lacked detail; one of the key metrics we wanted to measure was how long people spend on the app because that is a key indicator of customer retention and value delivery,” adds Medrano.
Jaime Medrano – NAVARRO LEAD ENGINEER
Tuenti
Tuenti Technologies was created in 2006 as a social networking service with over 15 million registered users. Tuenti has become a full Mobile Virtual Network Operator in Spain, owned by Telefonica. Tuenti offers communication services accessible anywhere by any device through its cloud based, unified social communication mobile app and the Tuenti website. The Tuenti social communication platform is also growing in Latin America where Telefonica Group is expanding the adoption in its customers across Argentina, Ecuador, Guatemala and Perù.
As its user base grew, Tuenti decided to find a platform that would provide a more detailed view of user behavior across the 6TB of data stored in its Hadoop environment. After careful evaluation, the company chose Vertica as its core analytics platform. The company has recently upgraded to version eight of this solution.
Vertica Analytics Platform enables Tuenti to manage and analyze massive volumes of disparate sources of customer data – from web logs and clickstreams to CRM and customer call detail records – to gain an in-depth, 360 degree view of customer interactions. Armed with deeper advanced analytical insights, Tuenti can improve customer web experiences, test and iterate products and services, improve marketing and campaign effectiveness, lower acquisition costs, and reduce churn.
“Vertica allows us to measure and aggregate multiple variables and metrics daily, pretty much in real time across the one million users of our mobile app,” continues Medrano. “It helps us prove correlations in user behavior, so we can focus on the most popular elements of the service.”
From an operations perspective, Vertica enables vastly improved reporting in terms of service usage while also ensuring compliance with in-country regulations.
“We can easily identify customers due an update and segment according to whether they rely on voice, VoIP, SMS or data,” comments Medrano. “That gives us much greater insight into our business and our consumers, so we can plan accordingly for future developments.”
Jaime Medrano – NAVARRO LEAD ENGINEER
Tuenti
Perhaps the most striking benefit of the Vertica platform is the speed with which queries can be answered and reports created. A typical example of counting unique users over a period of a year would have taken a full day or more to process using MySQL and Hadoop. Now, with Vertica this is performed in less than a minute.
One of the reasons why Vertica is so much faster is that previously Tuenti had to work with raw data in the Hadoop environment. This data is now compressed by Vertica and so annual data on unique users can be reduced from 1TB to just 50GB, enabling the company to have clear insight into how the app is used and adjust it in response.
“We measure how features are used and if they are not popular, we can drop them. Customized voicemail is one such example – customers weren’t using it and it wasn’t adding value, so we removed it from the app,” concludes Medrano. “Vertica is helping us to roll out three or four releases of the app every month because we can analyze the user data quickly and can respond to user demands. Additionally, this data is available to all of our 150 internal developers, not just the three-man analyst team. We call it democratizing data for better business.”
