On the 27th of June SAP organized an event for customers in ‘s Hertogenbosch (Netherlands) about SAP Conversational AI. We from Acorel attended the event as a partner. In this blog I would like to give a short introduction about SAP Conversational AI (CAI) and share with my experience during the day.
SAP Conversational AI
SAP Conversational AI (previously known as Recast.AI) is a product that is part of the ‘Intelligent Enterprise’ proposition/portfolio of SAP. Using SAP Conversational AI you can “automate your business processes and improve customer support with AI chatbots” as it says on the website. By using a low/no code modeling tool in the cloud, connected to your SAP solutions (On premise or in the cloud), you can easily create chatbots that carry out simple tasks (i.e. HR leave registration of order status lookup).
The day started with an inspiring introduction from Roxanne Arnts (Solution Advisor Artificial Intelligence at SAP in the Netherlands). Roxanne talked about why customers should be transforming, or should at least have plans to transform into an Intelligent Enterprise. One quick win, as part of becoming an intelligent enterprise can be the the introduction of a chatbot for certain tasks that are now handled by humans (i.e. customer service desk or HR support).
After Roxanne’s talk about why you should be transforming into an intelligent enterprise, Boris Andree of SAP (Solution Management Lead for Artificial Intelligence EMEA North) gave a nice overview of CAI and talked about why you should use SAP CAI as a tool to develop chatbots.
Business & IT Enablement Tracks
After the introduction the group of attendees were split into two separate tracks. The idea was that every attending customer would come in pairs of two to the event: one business person and one IT person. In the two separate enablement tracks both persons would be prepared for a hackathon in the afternoon.
IT Enablement track
The IT track was a hands-on workshop for the IT people to get acquainted with the CAI tooling. The basic components / structure of a chatbot where explained and a simple chatbot was build by following step by step exercises explained by Marcel Smits from SAP. By following this introduction, participants learned the basic components and mechanics of the CAI tool and were prepared to assist their business colleague with development of the chatbot during the hackathon in the afternoon.
Business enablement track
The business track was designed to help the customer with the following:
- Figure out chatbot use & business cases for their own organization
- Scoring the different use cases and figuring out the top use case
- Creating a value pitch for the top use case
After the exercises above, business people would be prepared to share their chatbot use case with their IT colleague and collaborate as a pair to create the chatbot during the hackathon in the afternoon.
Be Inpired: Lunch & Demo
After a busy morning a tasty lunch was served. During lunch I had the opportunity to show a demo app that we developed at Acorel. The app that I showed was a voice controlled chatbot that can recognize commands like “create opportunity for customer X” and create appointment with customer Y”. The app/chatbot can recognize these intents and will ask for additional variables that are needed (the value of an opportunity for example, and the date and time of the appointment). After gathering all information the app summarizes the transaction and saves it in the SAP Sales Cloud system.
After lunch the business and IT people was paired up to actually build a prototype of the chatbot. There was an excellent atmosphere at the SAP Experience Center. Everywhere you looked there was a team of two or three people behind one ore more laptops hacking away in SAP CAI. One of the teams even managed to connect their chatbot with an SAP Commerce system…really great stuff in a short amount of time. Everyone I talked to was really impressed by SAP CAI, mainly because of the fact how simple it is to use. It’s very intuitive and you don’t have to know anything about programming to create a simple prototype.
To determine a winner of the hackathon, every team had to record a three minute pitch. In the three minute pitch every team had the opportunity to present their use case and explain the business value of their chatbot. Besides the pitch itself, all teams also recorded a movie with a demo of their chatbot.
Judging & final presentatation by Skybuffer
After all pitches and demos were recorded, the judges gathered in a separate room to watch all movies and picked a winner. The price at stake was huge: a 50K investment from SAP in the form of the mobile innovation lab to develop an actual working prototype of the chatbot at the customer site for 5 days. While the judges were discussing a winner, a final presentation was given by Skybuffer. Skybuffer is a SAP Partner specialized in SAP CAI. They have already developed hundreds of chatbots connected to SAP systems for customers. Skybuffer shared their experiences throughout the years. A nice presentation to finish the day.
And the winner is…
The winner of the event turned out to be Stedin (a energy grid/network operator in the Netherlands). Their pitch/demo of a chatbot that could register power outage was picked based on the criteria of the judges. Congratulation to Stedin!
The SAP CAI event was one of the better one-day events I ever participated in. Not a boring day with only presentations but an excellent atmosphere with the focus on an actual end-result for all customers/teams by collaborating with each other supported by experts. Many thanks to the SAP team for organizing this event. See you next time!