New possibilities, never discovered before

Securities search system | Auto-evaluator

The goal with this project was to search in real time for securities, the best blending in with current portfolios of the clients. The main prerequiesite should be that the found securities could be arranged in descending order of likeability. Apart from that every security in portfolio should be automatically evaluated on the basis of assessments (i.e. short-, middle- and long-term forecasts) of the biggest players on the market. This approach should save hours of tedious work and increase precision of evaluation at least two-folds.

Here you can try the demo version of the product (with limited amount of securities):

Recognizing behavioral patterns of new-comers

The goal with this project was to automatically recognize behavioral patterns of users visiting different web-sites of a group of companies based on a source of entry, initial movements of mouse and scrolling activities (especially important on mobile devices). Patterns should be also automatically clusterized (using initial clusterization by the human expert). Such type of clusterization should help retain user on web-site and increase overall user satisfaction from getting information/help.

Here you can see a result of A/B test implementation on 10 web-sites within 4 weeks: