Big Data technologies brought to Aeroflot additional one billion rubles

Results

Big Data technologies brought to Aeroflot additional one billion rubles

Work with Big Data helped Aeroflot "to improve cardinally communications with clients and to attract more passengers". At the end of 2017 the airline earned from it in addition nearly 1 billion rubles, Aeroflot in the annual report published in August claims.

For increase in the entering flow of clients for the account for dynamic segmentation using arrays of unstructured data the company started commercial operation of Big Data technologies in December, 2016.

Big Data technologies brought to Aeroflot additional one billion rubles

Aeroflot called this project the first and only implementation of the systems of the class Big Data in the Russian transport aviation industry earlier.

A system allows to create the profiles of clients including including, a sex, age, marital status, interests, preferences and the history of flights, and taking into account individual data to offer more personified offers at the right time. A system also defines what channels to communication to use to this or that client.

In 2017 integration with external advertizing platforms Yandex Google, MyTarget was carried out it (Mail.ru) for start of a series of marketing campaigns on these platforms using dynamic segmentation of the customer base, Aeroflot in the annual report reports.

Further the company is going to develop the Big Data system for these purposes by adding of new data sources and optimization of models of machine learning.

In the strategy of development for information technologies for the Aeroflot group till 2020 approved in 2017 the particular emphasis on further multilevel use of Big Data, said in the annual report of airline is placed. Since 2018, Aeroflot is going to implement 2-3 new business cases a year in the field of Big Data and machine learning annually.

Project Progress

In 2016 implementation of the platform came to the end: the hardware and software system of the platform of Big Data was implemented, are carried out engineering and balancing and commissioning, the road map for 2017 is created.

In 2017 within the agreement of technical support the key business functionality of a system was finished: service of recommendations, client 360 and client's profile, impact assessment of communication channels, assessment of belonging of clients to a RFM segment. The cost of the mentioned works was 26.1 million rubles.^[1]

The project was executed using solutions of IBM: IBM InfoSphere Information Server, IBM Biginsights, IBM Streams, IBM Informix, IBM Big SQL.

Contractor Selection

The project started in December, 2015. The contract with Aeroflot for the amount of 148 million rubles "Technoserv Consulting" received according to the results of an open tender, having reduced the starting price of lot for only 5.8 thousand rubles. The companies tried to compete Interprocom with the winner and Integro Tekhnolodzhis. Performance of the contract was expected 260 days, i.e. it had to come to the end no later than September, 2016.

Structure of the solution

The solution created by a command of Aeroflot and Technoserv included several modules: "Client 360", Service of the recommendations, "Customer lifetime value" and "Tendency to a response and sensitivity to the price".

The module "Client 360" is directed to comparison and consolidation of data on passengers from different sources for finding of unique client profiles. Among such sources: daily unloading of records about flights and passengers from a booking system and reservation of air tickets, the unrolled characteristic of visitors of the website of Aeroflot from Google Analytics Premium and model of their behavior (search, viewing, armoring, payment, loading, link navigation, etc.) and also data on marketing communications (e-mail, the SMS and dr) and many other things. The family of the solutions IBM InfoSphere Information Server acts as the main tool allowing to perform data integration and to select unique client records.

The module "Service of Recommendations" selects alternative offers to the client for possible points of an aviatravel, based on historical data on its flights and the revealed correlations.

The module "Customer lifetime value" is an analysis algorithm of profiles of passengers of airline and consists of two blocks which purpose is measure definition of Customer Lifetime Value and belonging of clients to RFM segments (Recency, Frequency, Monetary — prescription, frequency, money — customer segmentation in the analysis of sale on loyalty).

The module "Tendency to a Response and Sensitivity to the Price" is an analysis algorithm of data on profiles of participants for determination of degree of susceptibility of clients to different communication channels. For this purpose in the module the possibility of calculation of tendency to a response of each unique client record depending on the individual frequency of responses is put. The module of the analysis of influence of communication channels is based on the algorithm allowing to fix participation of channels in communicating process, preceding purchase of each client (or a cluster) and also increase in number of new unique client records at implementation of media activity.

On a response of the customer, the project on creation of a system of intellectual customer segmentation provided to marketing department of Aeroflot the tool for understanding of distinctions and preferences of the audience, search of the hidden correlations, check of hypotheses. Using the analysis of Big Data and model of machine learning, specialists of the company will see off customer segmentation on a set of characteristics: their consumer capability, sensitivity to the price, potential return. The recommendatory systems daily calculate personal flight offers for each client. The address communications sent to clients in the fullness of time optimized marketing and, as a result, key financial indicators of Aeroflot increased.

Our project for Aeroflot became the first implementation of the systems of the class Big Data in a transport aviation industry of Russia. Besides, its implementation took less than a year that is the shortest time for projects of such class — Sergey Korneev, the president claims Technoserv. — Aeroflot performs global transition from simple sale of tickets to complex sale of a comfortable travel at the expense of customer attention, to parts, important for it: possibilities of transportation of mountain skis or availability of a special power supply. Enriching knowledge of clients from open sources, including social networks, an opportunity to perform trigger communications with the client, to react quickly to events in his life opens, realizing his requirements: moving, vacation planning, complaint to service of the company. Each of these situations can be quickly expressed by the relevatny offer raising a customer loyalty. And the knowledge of airline of the audience quicker reaction will be more complete, the loyalty and as result, profitability is higher. Behind it the future.

Notes