This flow starts with a customer calling in and being prompted to give feedback. Speech recognition will transfer it into text and analyze the tone and sentiment from it. Depending on the sentiment, the customer will hear that they will be sent a coupon after the call and have the option to stay on the line to talk to a rep. If they are happy, a message will play saying thank you for the feedback and lets them know that they are entered for the contest for giving their feedback.
Inbound Call – The Inbound Call trigger prepares the flow for a call.
Record Response – Record part or all of the call. You can select whether to only record the user, to only record your internal responder, or to record everything.
Speech Recognition – Convert the speech to text and analyze the sentiment and tonality.
Condition – Split the flow based on the sentiment of the caller.
- Use the sentiment variable that was generated from Speech recognition. Sentiment is on a –1 to 1 scale with -1 being a negative sentiment and 1 being a positive sentiment.
- All conditions must be true for the flow to continue through the top arrow’s path.
- If one or more of the conditions created are false then the flow will travel through the bottom path.
Transfer – Transfer the caller to another department.
From top to bottom, each port signifies what to do:
- On Answer
- On Hang-up
- On Answer
- On user Busy
SMS – Send an SMS.
Flows that are phone calls need to properly end the call. There are no inputs, so simply drag the action onto the board and connect it to the end of the flow.
Note: Atmosphere® SmartFlows does not currently save responses so any feedback collected will need to be stored in your own database.