Recommendation Engine

AI-powered Recommender API, providing the best product, media, event or game recommendations for a specific user based on user profile, behaviour and product details.

Subsequent user actions feed back into the learning algorithm, constantly improving it.

Usable by a consumer site to personalize their interface in app start screen, email campaigns or push notifications, users will be led to the content they will love resulting in better entertainment, less churn, and higher retention.


User Churn Alert

AI-powered user activity monitoring, and alert & retention intervention in real-time. Instead of reporting when a user has already lapsed, the service uses user activity and preference to predict upcoming user states, alerting the site & recommending retention mechanics.

This captures user churn in real-time, allowing consumer sites to retain users before they leave, in both those critical first 3 active days after signup, or over the long-term, increasing user life time.

In the mid-term future, this feature will replace much of the CRM process now done ad-hoc, saving on staff.


AI-powered CRM

AI enables internet businesses to connect deeply and personally with each user by moving away from impersonal CRM “bucketing” strategies, and embracing deep personalization. We use AI to map the user’s personal trajectory, and reinforcement learning agents to structure the optimal user retention strategy. This new technology, based on techniques used by AlphaGoZero, identifies dissatisfied users proactively and works to enhance and extend the site/user relationship. By constantly monitoring all such user incentive interactions and their effects, performance inevitably exceeds manual or bucket-based incentive approaches.


User Classifier

AI-powered real-time user monitoring & classification, plus an alert service via push notification or API. Fed with user activity and game preference, the User Classifier uses both supervised and unsupervised learning algorithms to classify users, e.g. detecting user behaviour patterns, and alert the operator.

Starting as support for operations and risk teams using rule-based systems to find defined user types, AI technologies constantly improve, in the long run identifying new user strategies.


Chat Host Assistant

An industry-first, the Chat Host Assistant provides AI-powered chat analytics for chat hosts & managers, to monitor and improve chat room dynamics in real-time, and A/B test and optimize different chat strategies based on composition of chat messages and chat room health.

It also derives subjects of interest to the users using AI, and pulls content from popular sources like twitter and news and presents them to the chat host for injection into the chat room, who can use as-is, reject or adapt.

Better insight and control into chat, the primary retention factor for users, allows the operational teams to react to changes in user sentiment better and faster, and to focus on handling specific user issues like retention. This leads to longer user sessions, more product purchases and in turn a longer LTV.


Marketing Content Agent

Reinforcement Learning Agents using your existing product offering and connect it to user state of mind (site, thread, posts, comments) to dynamically generate user-customized content, engaging the user on a personal level.

No technical integration required, operator involvement is only required for editorial guidance and acceptance testing.

>