As Artificial Intelligence is improving drastically, it causes bots to take up a major part of people's everyday life. AI enables computers to process language and even talk with people, in ways they never could before. Therefore, today's bots are able to help us order food, shop for clothes, save money and find restaurants.
Some bots are used to handle a variety of customer service requests, which would normally require a telephone call to a human agent.
Since most common types of bot are talkbots (used in the call center industry) and chatbots, Bot Report focuses on illustrating their performance and showing a variety of measures about them, providing to the top management a real-time analysis of how well the bot is doing, whether it is failing or succeeding in satisfying the customer’s needs.
Bot Report provides a variety of analysis where the managers can easily determine the bots’ highs and lows over a period of time. The solution is consisted of four different pages:
1- Global indicators show the performance of the bot in a given period of time. The point of this dashboard is to have an instant global picture of whether the business is going well or not. It provides metrics like rate of resolved and unresolved issues, rate of transferred, abandoned and prank sessions, average time of handling issues and so on.
2- Performance by Type of Session is an addition to the first page and it also shows global metrics about the bot’s performance. It shows the distribution of the number of sessions per session type, keywords occurrence, the number of sessions per category represented in a flow etc.
3- The Media type page shows an overview of the bot’s performance depending on the type of media. The bot can be either a chatbot or a talk bot, which is one of the filters available on this page. Besides the bot-type filter, there is also a date filter which enables the user to get a detailed report of the performance of the bot in each day/month or so. Since the talk bots are mostly present in the call center industry, whereas the chatbots are present on the social media networks like Facebook, Twitter, LinkedIn, Google+, we decided to go with these types of media and measure the performance of the bots here.
4- In the last, Capacity Planning page, measures and charts about the needed number of people are shown (with and without a bot), and predictions are calculated with the future planning metrics. It simulates increasing or downsizing the workforce, which gives a full understanding of the cost and resource impact whether the company owns a bot or not.
Bot Report turn dummy data about bots into a useful statistical analysis, empowering the top management to make meaningful decisions about investing in installation or improvement of the bot in their system. From these reports, the managers can spot some historical patterns and trends of the data, see comparison between some interesting KPIs and so on.