It’s common knowledge that when potential customers or leads contact a bank, the bank’s focus is now less on their request and more on getting them to buy something. Whether hidden or overt, the goal of any automated, semi-automated, or manually driven customer support center is to sell new banking products in every chat/call the bank receives.
Let’s imagine that someone contacts a bank via its chatbot to order a card (for example, Visa, Mastercard, or Local card). In line with the bank’s strategic goal, the chatbot solution must authenticate the chat user or capture them as a lead. This entails checking the user’s details via the authentication process and creating a sales quotation for the card issue in the system.
For the purposes of this example, we’re going to showcase a chatbot built with SAP Conversational AI and assume the bank is using SAP software as its digital core. Let’s also assume that SAP Fiori is deployed as the Internet banking portal, with users created in the SAP Fiori back-end system, where they are also given access to the self-services portal.
Of course, many initiatives are currently under way to develop an enterprise-level AI foundation on various platforms. To avoid presenting yet another unrealistic scenario built on a prototype that is far from market ready, we’ll base our example on real-world AI content – namely, Intelligent Decision Dimensions, an SAP certified add-on from Skybuffer. (Full disclosure: I also happen to work for the company.)
This solution can be deeply embedded into the customer’s SAP landscape (highlighted in blue on the architecture diagram below). Intelligent Decision Dimensions can be installed as an add-on to SAP on-premise systems. What’s more, the solution plugs into Cloud Foundry to get on-premise data via SAP Connectivity using the OData protocol.
To find out more about Skybuffer AI content built on the SAP Conversational AI platform, check out the following link.
With Skybuffer, users can simply start chatting to your SAP S/4HANA or SAP ECC system – which is the banking digital core system in our demo.
The Skybuffer AI Foundation package currently consists of 70 reusable skills. In the example that follows, this toolbox will enable us to present a realistic business scenario that masters the business challenge outlined above.
The following foundation skills are used to implement the banking business scenario presented here:
- Get First Name
- Get Last Name
- Get Phone
- Get Email
These simple skills are leveraged to obtain specific parameters from the user and to store them in the memory variables. A “Get Material” business skill has also been implemented, allowing users to choose the type of card they wish to order.
It’s important that the dialogue is controlled by memory and sentiment analysis. In every single skill from our AI foundation package, the user’s input is written to a variable, and the corresponding value is stored in the memory throughout the interaction. The data stored here includes the current topic of the dialogue and all the parameters that have already been asked for.
You can discuss several topics in the same dialogue and even move the discussion forward for all these topics simultaneously. The chatbot is clever enough to avoid multi-skills situations and move a topic forward based on the confidence value and current state of memory parameters.
Let’s now check out how the SAP Conversational AI platform can enable banks to generate business value.
Step 1: User Starts the Dialogue with the Request “Order New Card”
The user asks to order a new card. The first information the chatbot requests is the user’s phone number, which is a straightforward, unique parameter for user authentication. Here, the chatbot attempts to check whether the user exists and can be found in the banking digital core by phone number. In the present case, no results are found, so the chatbot suggests creating a new user. Now, the chatbot is attempting to capture a new lead in the banking digital core or CRM system.
We have now completed the initial step. In just a few seconds, we have found out that the person contacting us is a potential lead. So, in line with the bank’s strategy, the chatbot should now try to lay a foundation for selling them banking products.
Step 2: Create a New User ID and Provide Access to Internet Banking
The user goes through a quick-and-easy process to provide their first and last names along with their email address. The phone number is also a required parameter when creating a new user. This variable is already stored in memory from the first step.
Let’s assume we have a happy flow, and the user follows the chatbot:
You can see just how quick and convenient the process is. A lead has now been created in the digital banking core system, and we have captured their email address and phone number. The Internet banking user ID is sent to the email address provided. This enables us to easily track whether Internet banking login has been performed and mark the email address as verified or not in the digital banking core. The following figure shows an example of the Internet banking login information email:
With the username and password we have provided, the potential lead can now access the bank’s products via the Internet banking solution.
Step 3: Verify Contact Details and Create Bank Card Sales Quotation
The chatbot – or perhaps we should call it an Intelligent Assistant in this rich business scenario – continues to pave the way for sales of bank products and now tries to verify either the lead’s email address or phone number:
During the email verification step, a session key is used. The following is a sample email we’ve created for this business scenario. Notifications of this kind are a standard feature of the Intelligent Decision Dimensions solution:
We’ve now completed our scenario. In a dialogue lasting only a few seconds, we’ve delivered the following business value:
- We have captured a new lead in the banking digital core with their verified email address, enabling the call center team to take this contact into operation, complete the card-ordering process, and even sell other bank products.
- We have provided an Internet banking user, giving our lead immediate access to the banking tools for ordering new bank products in self-service mode.
To round out our scenario, the following screenshots show the digital banking core transaction data created by the embedded Intelligent Assistant (our next-level chatbot).
Thanks to our Intelligent Assistant, banks can now sell their products in fully automated mode. In the example above, we showed how users can order a new bank card without talking to a human operator or visiting a branch. The process could hardly be simpler: all users have to do is interact with the chatbot, complete the validation procedure, and select a bank product type. While this is going on, a quotation is created in the digital core system. Once the bank has contacted the lead, requested additional information (if this was not possible during the chat session), and discussed the agreement for bank product delivery, an operator can change the status from quotation to order.
Ready to transform your customer experience with SAP Conversational AI? Please contact our team if you’re interested in knowing more about how to integrate chatbots in your business. You can also see how Groupe Mutuel is improving customer service and satisfaction with a chatbot that lets customers ask questions 24×7.
What is Skybuffer?
Skybuffer is an international SAP implementation and development company founded in 2013.
Since 2018, Skybuffer has been focused on developing SAP Conversational AI-based new user experience for SAP on-premise systems (SAP Business Suite and SAP S/4HANA), offering business users opportunities to work with SAP systems via text (chatbots) and voice-enabled channels.
Skybuffer hosts a public innovative demo portal where anyone can try out the new SAP Conversational AI-based user experience to get an idea of how chatting or voice-talking to SAP systems reduces stress, increases productivity, and facilitates access to SAP on-premise data.