Beyond just showing customers how to use a voice agent, a multi-agent product will have to help customers find the agents that are available to them, choose among them, and explore what they are capable of. What is the most effective way to show customers how to get the most out of the voice agents on your product?
Because voice is a largely invisible modality that relies heavily on customer recall, it is important to be clear about how to engage with multiple voice agents, and to get customers excited about the possibilities.
The registration and education flow, available to customers either during the product’s out-of-box experience (OOBE) or afterward, might be a good place to introduce your customers to available agents. Including additional discovery mechanisms is another great way to encourage customer engagement.
Much of the customer education and initial settings choices are handled in a product’s out-of-box experience (OOBE). The OOBE setup flow can cause cognitive overload for customers due to the many decisions they must consider. For customers who choose not to complete the voice setup during OOBE, or who want to change their settings, offer them an easy way to access those settings at any time, for example in a companion app.
We recommend that a companion app should allow customers to:
Locate voice and agent settings in a single, easy-to-find area of the companion app. All products should provide the following settings and controls from within their device’s companion app:
The device maker should supply the customer a place to manage their voice agent- specific settings.
Agent makers should allow customers to change settings and preferences specific to their voice agent.
In multi-agent experiences, customers should always know what agent they are talking to. Agent attribution may be explicit (e.g. colors, logos), or it may be indicated by differentiated visual and sound cues. Attribution can also include both personality and behavioral characteristics unique to an agent.