Intro –
24 months of product writing with multiple iterations, implementation cycles and after an intensive round of Bot training they finally unveiled our own Streebo Enterprise Bot. This Bot would provide a chat and voice interface to our complete Intranet. Existing Intranet has over 15 applications and soon going to over 25 applications (web and mobile). Now that the new Enterprise Bot is in place – what will be different ?
What will the BOT do?
Our Enterprise Bot (yet to be named) is powered by Artificial Intelligence (IBM Watson) and it extends the Intranet from Web and Mobile to a Conversational Interface – The Bot is stationed on our web and mobile app and our team members can now give chat and voice commands to the underlying applications – In the Old World, our Lab Associate had to log in to the Intranet and go to upto 3-4 disparate systems to do a multitude of tasks throughout the day. In a typical day, the associate would maybe start out by checking their pending items in the “Work planning System” – next maybe go into “Leave Management System” to apply for leaves and finally into our “Bug Tracking System” to close assigned bugs. In the new world our Lab Research Associate simply invokes voice/chat commands like “complete pending tasks” or “apply for leave” and finally “close bug number so and so” or “show me pending bugs assigned to me” and the Enterprise Bot goes to the relevant system of records and fetches data or updates the same. Same applies to various employees in other departments. In the old world, the Sales Reps would go to the CRM to check or update data and notes. They would review cadence reports. Maybe file for expenses using the “expense management system” and maybe open a travel requisition using the “travel portal” – in the new world they again interact with the same legacy systems but using chat and voice commands like “file a new expense report” or “update the CRM record for so and so customer” and finally “book new tickets for an upcoming sales trip” – the bot then guides them and collects requisite information or fetches the same from the legacy backends. This will allow our employees to be more productive and will result in improved adoption of our Intranet.
How was it built?
The Enterprise Bot was assembled using a combination of IBM Watson & Cloud, HCL Digital Experience Technology and Streebo DXA. While Intents were put into IBM Watson Assistant, the guided conversations including integration with system of records was done using the low code no code tooling for Watson assistant and i.e. “Streebo Bot Builder”. This allowed us to leverage a combination of public-private cloud architecture to roll-out our solution which was secured yet flexible and optimized for ongoing costs. Thanks to the drag and drop interface and point and click integration available in the Streebo Bot Builder; Junior Research Associates, Freshers and even interns could completely assemble the entire Bot with little or no training. The IT can centrally manage the security and access control from the Central Conversational Platform interface to ensure that security remains tight –
Party Highlights –
The highlight of the party was each team (built across various departments) had built their own menu and in many cases even prepared their own food. There was a Bot designed to even order the food and it was assembled and trained in a few hours. We could even order the food through the Whatsapp interface. The Spicy Mocktails and shots were the key highlights – so were the loud noises and cheers that still ring in my ears as we unveiled the new Artificial Intelligence powered bots – now on to the next phase wherein we watch the analytics everyday, keep training the bot to make it incrementally smarter and watch the adoption grow … onward and forward …