News

Google's chatbot analytics platform launches

Anyone can use it to get a deeper understanding of how their chatbot works

17 Nov 2017

Google has launched its chatbot analytics platform dubbed Chatbase, providng devlopers with more insight into the behaviour of their bots.

Chatbase has been designed as a cloud service which allows users to analyse and optimise bots and is completely free to use. It came out of Google's incubator for early stage products called Area 120.

Ofer Ronen, the Chatbase team lead, wrote: "With Chatbase, bot builders can quickly understand interactions in order to increase conversion rates, improve accuracy, and create a better user experience overall. The service is free to use, easily integrates with any bot platform, and works with any type of bot, voice or text."

The service uses Google's machine learning capabilities to automate the identification of bot problems and opportunities. Companies which have adopted this technology already include Ticketmaster and HBO.

Viber, a messaging platform with over 900 million users across 193 countries, said: "Chatbase gives us powerful analytics capabilities for bots. We increased query volume by 35% for a popular stickers bot by optimizing queries with high exit rates.

"Chatbase has been immensely helpful in improving our bot. Instead of combing through logs, we rely on its machine learning capability to help prioritize required optimizations -- saving precious time that we need to focus on building new features."

Chatbots the next "epoch of tech"

Chatbase contains a number of features including the analytics platform which updates multiple times an hour and shows active users, sessions and retention.

Bot accuracy can also be optimised faster, with Ronen noting: "Our machine learning model clusters user messages that aren't handled well, identifies opportunities to answer more requests, and offers paths to easy optimizations that address both (currently available via an Early Access program)."

The service also auto-generates a data visualisation of conversation flows across sessions allowing users to see which path customers take and where they end up during their interactions with a chatbot.

Image source: Chatbase