Conversational Hybrid support:
The best of both worlds
We understand that when it comes to product support for physical products, clarity and accuracy are paramount, but so is a natural, engaging conversation. With our new Conversational Hybrid Experience, you no longer have to choose.
This update seamlessly blends the precision of your curated answers with the dynamic flow of generative answers. It means that when a customer asks a question, our smart AI algorithm intelligently decides the best way to respond. It can:
- Surgically inject curated answers directly into the conversation when specific information needs to be presented visually and/or with structured forms, or when it’s important that the customer follows a specific process exactly as specified.
- Generate natural, conversational responses for broader inquiries about your products, processes etc.
The result is a unified, engaging customer journey where every turn - curated or generative - feels like a natural conversation.
Now let’s look closer at the key elements that make Hybrid conversational.
Typing indicator
It smoothes the wait time for AI’s answer and makes interaction feel alive
Answer chunking
We now break complex answers into logical chunks that unfold gradually - so customers can digest them better. Goodbye, “walls of text”!
AI face and a name
You can now choose to show a custom-branded avatar and concise name of your assistant, with the AI disclaimer - to set the expectations on answers generated by AI.
Pre- and postambles
These are contextual opening and closing lines for curated turns, that engage users in trying a new solution, choosing the next option or taking a next step.
A short preamble is useful to confirm the user’s intent (“If your grill isn’t heating properly…”) and explain multiple suggestions.
Postambles are hints on what to do next - clarify, rephrase or add more details on the issue.
Multimodal-in-browser:
Making the impossible possible
Imagine your customers speaking directly to a knowledgeable assistant right on your website or app – not just typing but speaking, sending pictures, and receiving real-time, personalised assistance. This vision is now a reality with our Multimodal-in-Browser, currently in beta.
It dramatically streamlines complex processes and the ability to solve problems that were previously cumbersome or even impossible in a single interaction.Consider filing a warranty claim. Traditionally, this might involve multiple calls, follow-up emails and attachments. With Multimodal-in-Browser, a customer can:
- Speak their issue directly to your website/app assistant
- Upload a photo of the faulty product right away
- Send their proof of purchase – all within one continuous, fluid conversation.
This slashes the time to complete such complex tasks from “days” to “minutes”, delivering an incredibly pleasant and efficient customer experience.
Voice Assist:
More human-like and scalable
Beyond these major releases, we’ve also rolled out a series of enhancements to Voice Assist. These updates are all about making interactions smoother and more effective, ensuring your voice assistant truly sounds and feels like an extension of your brand. At a glance:
- Better turn-taking. Voice Assist now understands the rhythm of real conversations better than ever - it knows when to wait, when to listen and when to speak - making conversations feel more responsive and intuitive. Customers can easily interrupt or change direction mid-sentence without friction, and responses come through faster, thanks to an upgrade to our underlying speech engine.
- Better pacing. When it comes to guiding customers through multi-step tasks, Voice Assist now breaks things down more clearly. Instead of delivering dense answers all at once, it guides users step-by-step, with natural pauses and checkpoints. This makes it easier for people to follow along and stay engaged throughout the call.
- Custom tone and behaviour. You can also tailor the tone and personality of your assistant more precisely. Whether your brand voice is friendly, neutral, or formal, you can now configure how the VA sounds and behaves. Choose from a wide range of voices, adjust pacing, control escalation triggers, and define exactly how key terms should be pronounced - so your assistant sounds like an extension of your team
- Multi-product linking. For companies with broad product catalogs, Voice Assist is now equipped to handle more complex structures. Using new flow linking capabilities, the assistant can jump between different flows - like moving from a general product category to a specific item, then back again 0 without losing context or requiring restarts. This helps maintain both precision and coverage across multiple product lines.
- Vague questions handling. We’ve upgraded how the VA understands and routes queries. A new intent recognition system enables it to ask clarifying questions when the customer input is vague, so we get to the nature of their issue and give precise advice
- More multilanguage. Finally, VA now supports an expanded set of languages, including Bulgarian, Czech, Greek, Hungarian, Lithuanian, Latvian, Romanian, Slovak and Serbian (in addition to 30+ languages already supported). For many brands, this unlocks the ability to support customers in their native language - even when no support agents in the team speak it.
Have ideas?
Let’s build the future of support together
We’re dedicated to creating product support experiences that set the gold standard for brands with physical products and devices.
If you have feedback, challenges or ideas that can help us push the bar even higher, share them directly with our product team using this form.
