Best Practices: Speech-based IVRs
Speech recognition has been playing an important role in improving the adoption of Interactive Voice Response (IVR) systems across industries. Companies have been using speech-enabled IVRs to provide natural, efficient and faster, customer service.
Although deploying speech recognition can be an intimidating task. If not done properly it can cost you a lot and will not add the value desired. So here we are to help you with some of the best practices you can leverage while deploying speech recognition technologies in the Contact Centre:
- Identify the Need:
Speech recognition for IVR is a relatively new concept when compared to DTMF. So it becomes more important to analyze your requirements thoroughly. It pays to plan well. Start with identifying answers to the basic questions like “ What are our biggest call drivers?” or “What information do our customers need the most?”. Not everything needs speech recognition.
You should not overcomplicate your call flows with the introduction of speech recognition and replace questions directly. In fact, you should focus on revisiting the entire flow and see how you can expressly make the process more efficient using speech recognition.
Contact Centre data and interviews with agents can help you to gain insights into caller behavior, therefore simplifying the answers to the above questions. Prepare thorough documentation for this phase which will come handy in the future as well.
- Flow Designing:
Next comes designing the flow of IVR conversations. IVRs with speech recognition reduces menu options and free the customers from the hard efforts of punching number. The key is not to design conversations but guided call flows.
Here are some key practices to take care of while designing an IVR flow:
- Establish an objective: understand what the IVR is supposed to accomplish.
- Keep your questions close-ended. Always help the client get close-ended questions For e.g:
“What is my account balance?”
“What time is flight AI 101?”
- Flows should not be defined to get monosyllabic answers. Design flows to get proper questions with intents (verbs) as responses, rather than just yes or no. Rather than “will you like to know your account balance?”, get the IVR to announce “In order to know your account balance say, what’s my account balance?”
- Avoid using jargons used by the company but are foreign to the customer.
- Keep it Short and Simple:
Follow the KISS strategy. IVR conversations should not have more than two or maximum three levels of interactions. As with the increasing level of interactions we might lose perspective of the conversation with the customers.
- Personalise & Contextualise the Flow:
Along with shorter conversations, every individual customer always seeks personalized attention. It is necessary for us to make the customers feel special by providing them meaningful and relevant conversation. One simple and effective way to achieve this is by starting to greet them by their name, for e.g.
“Hi Priya, Welcome to Phonon Communications”.
Follow this up by contextual questions, so in case Priya is a caller to an airline who has a ticket, announce
“Hi Priya, Welcome to your favorite airline! To know the status of your flight say FLIGHT STATUS, to Check-in say CHECK-IN, for an existing booking say CUSTOMER SERVICE or say BOOK A TICKET to make a new reservation.”
Phonon’s skill in NLP and Speech Recognition:
Phonon’s Intelligent IVR provides the caller multiple user input options. It has the capability to process both DTMF and voice input from the caller. It enables callers to use everyday language to solve their queries. Supporting close to 12 Indian languages, our IVR combines natural language processing and machine learning to ensure higher accuracy in detecting the customer’s intent and responds intelligently.
The seamless integration of our Intelligent IVR with different global AI platforms like Google Cloud, AWS, Azure, IBM Watson gives the users freedom to be AI/ML platform-agnostic.
As a value addition to our consultative exercise, we also provide industry-specific ML consultation to bring inflow level automation in customer service flows.