How to reduce immediate human escalation
Reduce immediate human escalation and improve automation. Learn how to reduce the number of users who ask to speak to a human without trying the virtual agent first.
VA Conversation Design
-
-
Tracking important flows with goals
In this path, we'll talk about goals. Goals track flow usage. They generate statistics about how many people go through (or complete) your flows, and can tell you important things about how your users are using your virtual agent.
-
Convince your users: not the least, but lesser effort
When we create a VA, we want nothing more than our customers to engage with it. This isn't always easy - sometimes they need a little convincing, or a nudge in the right direction. We show that you need to...
-
Seven simple issues and how to fix them
In this session, we show 7 common issues with conversation flows, why they are bad for your customer's experience, what you can do to avoid them, and how to fix it if you have them yourself!
-
How humanlike should your chatbot be?
Intuition and research alike tells us that human likeness helps to make virtual agents (VAs) more trustworthy, but when a VA fails to provide help - does it really matter how human like it is? (Spoiler: it does not!)
-
Interview or usability test checklist
A step by step document detailing the things you need to think about when planning an interview / usability test, including a downloadable (and printable) version to bring with you in real life.
-
A/B testing, a look at new tools
A/B tests allow you to test new flow designs & to gather data on which performs best. Join us to take a look at how it works, how to set it up and talk about the best A/B test practices.
-
Tips & tricks for a good interview
A document with tips and tricks for when you've planned an interview/usability test and need a little reminder, including a downloadable (and printable) version to bring with you in real life.
-
How do buttons influence UX?
Do you know when to use buttons (and when not to)? Let yourself be informed by what research has to say - including the latest insights from the Conversational User Interfaces (CUI) conference 2023.
-
Introduction to the LLM action
In this session, we will show you the magic of the LLM action - one of v12.1's biggest new features. You'll learn how to set one up and what kinds of things you can start handing off to LLMs within...
-
Using humour for conversational repair
Humans use humour to deal with awkward situations and repair mistakes. So should our VAs use humour, too? This article takes you through a research paper by Clausen and colleagues, published at the Conversational User Interface (CUI) conference 2023.
-
Leave the persona for last
What to prioritize - designing a good personality, or building reliable answers? Let's see what your users expect, and you'll be more comfortable leaving the persona design for last. This article is based on work from Makany and colleagues, published...
-
Show your understanding: paraphrase
Your VA should show that it understands your users. But how should it do that? This article takes a look at the study from Papenmeier and Topp (2023), published at the Conversational User Interfaces (CUI) conference 2023.
-
Inclusive VA design: How to design for all users
In this article, we dive into the concept of inclusive design: Designing for vulnerable users. Regardless of your virtual agent (VA)'s use case, you will have vulnerable users, and it is good to think about how to design your VA...
-
Building virtual agents with Gen AI
Generative AI helps us build smarter virtual agents, faster. But it's not without its challenges. We combine large language models (LLMs) with natural language understanding (NLU) to provide guardrails that make them safe for enterprise use.
