Aeroflot deployed the Platform for work with customer requests based on the predictive analytics technologies Big Data

At the end of 2017, PAO Aeroflot deployed a unique Platform for dealing with customer requests, and was the first among the transport industry companies who used the algorithms for predictive data analytics.
Integro Technologies Company implemented the project, which is part of the RAMAX Group. The project used predictive analytics technologies based on Big Data using IBM BigInsights and IBM PureData platforms.

Increasing loyalty is impossible in the absence of a well-functioning and prompt reaction to the feedback, complaints and information requests left by the passengers coming from different communication channels (including social networks). Loyalty is enhanced if the company is open to comments and suggestions, and customers are confident that their wishes will be heard and accepted.

To increase the efficiency, speed and quality of processing client requests, it was necessary:

  • Create a single platform for processing calls from different communication channels, their intellectual classification and routing;
  • Provide a tool that allows online monitoring of key trends in social networks, including feedback about PAO Aeroflot from opinion leaders (popular bloggers, journalists, etc.) and if necessary, use all available communication channels to preserve the company’s image.

After researching of the market for solutions supporting Big Data technology, PAO Aeroflot specialists concluded that IBM BigInsights and IBM PureData are the most functional and flexible solutions and meet the needs of the airline, and Integro Technologies has established itself as a reliable partner for customers from the civil aviation sector and has all the necessary competencies to work with these platforms.

Within the framework of the project, the fundamental functions of the system were designed and implemented, allowing flexibly configuring and optimizing the algorithms of its operation. 

Firstly, the process of monitoring of the references to PAO Aeroflot in social networks, tracking their tonality and the dynamics of changes following various information events was introduced. Analysis of reviews from social networks was conducted on an array of the entire Russian-speaking segment beginning from January 2014. The data obtained was used for machine learning, which allows real-time determination of the category and tonality of a specific reference, helping responsible employees correctly determine the priorities for responding to messages in social media.

Secondly, on the basis of the accumulated historical data, mathematical models were constructed that allow, when receiving a message from the client via any communication channel, automatically (with high probability) determine the topic of the message, as well as the subdivision responsible for handling and solving the problem. This approach has reduced the use of resources to analyze and reassign the requests received. Moreover, in the process of work, the system analyzes the “fresh” data and continuously learns to improve accuracy. 

By combining these two blocks, it was possible to implement the end-to-end process, starting from receiving a message in a free form in one of the social networks, and ending with providing a full reply for the Customer, and detailed analysis of the reasons for the occurred and unified reporting on messages for making managerial decisions for the Company’s management. 

The uniqueness of the project

At the current time, an assessment is being made of increasing the efficiency of the work of employees, it can now be fixed that the system corrects about 5% of the error topics of messages and independently determines the topic in 85% of cases. 

Due to a prompt response to the most resonant news occasions, the loyalty of potential and existing clients of PAO Aeroflot has significantly increased.

Integro Technologies Company implemented the industrial implementation of Big Data technology for automation and integration of processes at the interface of traditional communication channels with clients and social services. 

At the same time, the results of the project showed the unconditional efficiency of the technology, which is confirmed by the Customer’s plans to expand its use.

Kirill Bogdanov, CIO of PAO Aeroflot: “The basis for all current and future changes in the approach to working with customers are technologies for large data management. Thanks to Big Data, PAO Aeroflot has new opportunities to conduct efficient business.
This project can be considered definitely a breakthrough of the year. CRM projects already exist in the world aviation, but all of them are without a global collection and analysis of data from social networks. However, the client’s behavioral profile is accumulated right there. As part of the work with clients, Aeroflot took a step forward, building an incredibly large volume of sources of feedback from customers to obtain the most accurate and reliable picture of the relationship, timely response to complaints and messages. All this leads to an increase in the level of loyalty, and ultimately to an increase in the sales volume of the company.”

Source: Global CIO portal
* — must be filled in