Innovation of the month: Tips in Sberbank Online
Efma’s innovation of the month award for March goes to Sberbank for its artificial intelligence based service called Tips. Sergey Komarov, the bank’s product owner, tells us more.
Traditional personal finance management (PFM) services involve a retrospective – as a rule the user can look at a history of their expenses, group them according to categories and then make their own conclusions. But this sort of analysis demands time and effort, and needs to be performed over and over again. As a result (and statistics prove this) few people able to regularly analyse their expenses.
With this in mind, we wanted to help our users to change their financial habits by achieving true automation of PFM. We wanted to switch the focus from the past to the future.
In the banking industry, machine learning is usually spoken about with regards to internal banking systems such as risk assessments, security and marketing. While Sberbank also uses machine learning in those areas, we wanted to add to this list and employ machine learning for the client and to benefit the client – to make online banking even more useful by creating a true assistant in the world of personal finance and everyday life. So we created a concept for a recommendation-based service, which uses algorithms to offer personal tips that are relevant to specific users.
Tips is a part of the Sberbank Online mobile app, especially for private customers. It uses artificial intelligence (AI) and help users save money, time and effort. Tips analyses the user’s financial behaviour and then leverages a set of machine learning models, including experiments with neural networks.
This service is a part of the deep evolution of personal finance tools. It makes it possible to break free from traditional analysis of expenses, look to the future and change one’s financial habits. It has been created at the intersection of digital assistants such as Siri and financial services like Mint and offers the advantages of each: proactivity, analysis of financial behaviour, machine learning, provision of recommendations for everyday tasks, and not just money-related ones.
Tips analyses user data (financial behaviour including spending and savings and profile data) and makes conclusions based on tree of life situations. It uses a set of machine learning models, including experiments with neural networks.
We can roughly divide tips into three categories:
-Complicated (using machine learning) – are shown to a client, provided there is a great probability of some sort of event happening in their life. For example, a user is planning a new addition to his/her family – the assistant gives the user a check-list with a documents and addresses of the organisations where they should be issued during the first month of the baby’s life.
-Middle-level tips (trigger advice) – shown to the client after certain types of activities. For example, a user pays to rent a car abroad – the assistant suggests that he/she should re-fuel using stations on secondary roads, because the fuel is about 5-7% cheaper there than on primary roads.
-Simple tips – advice occurring from time to time, with easy targeting associated with specific dates. For example, on a weekly basis the user receives interesting facts and statistics about his/her expenses. This is an entertaining way of reminding him/her to think about his financial behaviour.
There are serveral benefits of our Tips for Sberbank Online users. First, the user can easily give feedback that allows us to target recommendations more precisely. It’s not just about advice – many tips end with the script to go to a website or moving within the application with filled-in fields. Thus, the recommendations are more likely to lead to concrete action and help the client.
We are steadily expanding to different branches of life scenarios, rather than randomly moving in different directions, so you can be sure that in certain areas the service will provide its recommendations. Now we are working on functions of socialisation for the Tips users the opportunity to share them with each other; the ability to share information about a sales point in a particular institution – they will come to a friend also in the form of advice (e.g. “Look, they have great soups here! Definitely come and check them out, you’ll love it”) after the transaction.
Eight Sberbank departments worked to create Tips (to deliver the concept, develop the product, generate content, create algorithms, support clients, etc).
In February, Tips became available to 20 million active users of Sberbank Online. Currently the service is used by around 200,000 users daily. We are gathering feedback, assessing existing recommendations and preparing new ones. There are 50 tips available now, in a month there will be 100, and this number will continue to grow.
Looking ahead, we intend to expand functionality, experiment with new possibilities (USG, clarifying questions etc) and add third-party content suppliers.
Submit for the Efma Accenture DMI Awards 2017 before 7 April and compete for the April Innovation of the Month (Innovation of the Month is reserved for Efma members)