Have you ever had a bad experience with a chatbot? There are over 300,000 bots on Messenger so the odds are pretty good that you’ve chatted with at least one that wasn’t quite ideal. Maybe it got lost and wasn’t able to finish the conversation. Maybe it gave you the wrong weather forecast. This one negative experience might have negatively affected your perception of chatbots.
A good conversational design can do wonders for your business. Chatbots are excellent customer service tools that can improve your sales, provide a unique customer experience, and automate some of the most tedious tasks. However, to unleash their full potential and have better control over them, you should be able to understand how conversations between them and your customers work.
What is conversational design?
The concept of conversational design is about looking at human conversation as a model for all interactions with digital systems. Using the principles of what makes everyday human interactions productive, it’s possible to create a better and more natural dialogue with systems.
The conversation is how humans interact with one another — any two strangers who speak the same language can have a conversation using this familiar interface.
Even if it’s occasionally awkward, or you don’t fully understand all the inner workings of another person, there is enough shared understanding about what’s expected in a conversation to efficiently communicate. The goal of conversational design is to learn from human conversations to make digital systems easy and intuitive to use.
Principles for conversational design go beyond voice assistants and chatbots — UI, web design, and even print design can all can be more conversational. Intentional language choices can make digital interactions feel like they’ve been designed for humans, by humans.
Why do we need conversation design?
Unlike mobile apps, conversational interfaces give users the freedom to say anything. There is a huge variety of ways that people ask for the same thing. Very often they also have a problem with naming their problems at all. This makes it very difficult for a chatbot to solve every problem a user might have.
The aim of conversation design is to map out what users might say and teach a chatbot to react in a way that helps users easily achieve their goals.
3 Pillars of a Conversation
Before we create our first dialog, we should take a quick look at 3 general principles that distinguish successful conversations and dialogs.
In linguistics, the cooperation principle is used to describe how communication participants normally behave – in other words, how we can expect the users to act when they use a conversational interface. We assume that a user wants to lead the conversation to their goal with their messages – and the user also assumes that the digital assistant is also goal-oriented. Every message is interpreted as relevant in this context and thus harbors the chance that the conversation can be much more efficient, but also the risk that misunderstandings arise.
As a rule, dialogues need at least two partners who mutually contribute – which is called turn-taking in linguistics. In natural conversations, we don’t worry about handing over the speaking role. We almost automatically take over after questions, speaking pauses or filler words.
When designing the dialogue, however, we have to use this very consciously in order to put the user in the role of the speaker. Usually the user can only give one answer, then the system takes over again. This form of dialog guidance – short answers from the user followed by an interpretation and answer by the digital assistant – is particularly challenging if, for example, the user was interrupted too early or sent a message incorrectly. In dialogue design, correction options and “going back” must be taken into account in advance.
Even when it comes to voice input, we now know exactly what the user said – thanks to Advanced Speech Recognition. But interpreting what it means is still a challenge. This interpretation is also difficult because the relative context of the conversation has to be taken into account.
It is challenging because the same words in different scenarios can have different meanings. For example, the word “tape” could either refer to recording equipment, packaging material or the action of recording something or gluing something together. The context is decisive.
But not only the meaning of words, but that of entire statements is context-sensitive, i.e. dependent on the environment or situation of the user. This is precisely why it is important in conversational design to always develop dialogues and logic processes as close to the user’s reality as possible.
8 best conversational design practices
Create a persona of your ideal customer
The first rule of conversation design is to start with the audience and not the design itself. To design great chatbot conversations, you should remember who your customers are.
Create a personification of your target clients and try to describe them in detail. Maybe your persona is a vegetarian woman in her early 30s who lives in Germany? Or a technophile from India?
It all matters, because your customers can vary in their shopping behaviour, preferred everyday communication channels, level of expertise or even degree of English proficiency.
Have you ever tried to insert yourself into a conversation and feel completely lost? It’s confusing at best and leaves you discouraged and unwelcomed.
Before designing your conversational interface, get to know who you may be interacting with, the context of the conversation, and what the users may want to get out of it. Purpose provides direction which can help lead to more fruitful engagements.
Having a clear goal in mind is a core principle of interaction design. People have goals when they interact with digital systems, services, and products — whether it’s checking a bank account balance, asking for help, comparing vacation spots, or looking up an unfamiliar word.
User goals and needs should be explored via user research as part of the holistic design process, and they’re key to designing a successful interaction. If you don’t know what your users are trying to do, how will you know when they’ve done it successfully?
A successful interaction helps both parties — customers and organizations — meet their goals.
Choose the Voice That Matches Your Brand Personality
When you know what your goals and who your customers are, you can finally think about the things you want the chatbot to say. Make the users feel comfortable during the conversation. You can use everyday language and informal style or go for something more reassuring and professional. The voice should reflect the values that your business embodies.
Your brand probably has a unique visual identity already – you use a logo and your promotional materials are made in a specific visual style. You can do the same with your brand voice and add a chatbot to your brand book. The conversation design of your chatbots should be distinctive and linked with the character of your business.
Be careful! Chatbot messages are not similar to advertising texts. They are interactive, and users navigate through the conversation to complete specific tasks or find out about something. So, it would be great to add your branding elements, logo, and futuristic fonts to navigate them better. Remember, that in a way it is a part of your website interface and its primary function is to be easy to use. If a chatbot’s “personality” draws too much attention, it may hamper user experience and become irritating.
The equivalent of “reading a room” to guide a conversation, the more context-aware a system is, the more conversational it can be.
When you’re searching for hotel rooms available tonight in Seattle on your phone, for example, you don’t want to see rooms available in Boston, or deals on hotels+flight+rental car packages — it’s clear that you are already in Seattle and need a place to sleep.
Overly automated messages or recommendations that don’t add value to your users can be damaging to the overall experience. The more a system can respond to contextual cues, the better it will be at having what seems like a natural conversation and not leaving users feeling stranded.
While devices can automatically provide information such as a user’s location and time zone, additional insight into what users need and expect throughout the different phases of their interactions helps inform solutions that feel like they’ve been designed for humans.
A successful conversation offers reliable information and prevents confusion. A truthful conversational interface delivers content that is expected by users then. Your users should always know what they will find behind the link, buttons, and cards your chatbot is sending. You should never lure users with misleading copy like “Get started” just to make them subscribe to a product or a product design they know nothing about. Such methods ruin the conversation and lowers trust.
To inspire trust your chatbot must also deliver reliable answers. Remember to update your chatbot content whenever anything changes. Have a chat with it from time to time to pick up possible mistakes or outdated information. If your chatbot lets users down by providing incorrect responses or data, they won’t trust it in the future.
Being polite doesn’t only mean using proper language. It also means respecting customers’ time. Erika Hall said that “polite designs are those that meet business goals without interrupting the customer’s pursuit of their own objectives.” (Conversational Design)
When a customer starts a chat to solve their problem, don’t push offers and discounts on them. Anticipate customers’ needs and help them resolve their issues quickly and without distracting them. If you really want to use that occasion to ask a customer to fulfill a survey for you, ask politely for it only when a customer completes their task.
Map out Possible Conversation Outcomes
Hitting a dead-end is one of the most common conversation design traps. When your customers don’t know how to respond or what to do next, they may become frustrated. Or, even worse, they may think that it is their fault and that they are to blame for not being able to use your chatbot correctly. These negative feelings can discourage them from visiting your website again.
Anticipate for Conversational Error
People don’t always understand what others are saying. Making mistakes is in our nature and very often people need to resolve misunderstandings while having a conversation to make it effective. The same is true for chatbots. It’s normal that they don’t always understand the user. The point is to teach them to quickly resolve misunderstandings the way humans do.
It happens in everyday conversation - maybe they misheard you (no match), get distracted and don’t respond (no input), or provide an answer that doesn’t necessarily fit your question (system error). They’re hard to expect but you can hide errors by designing conversation repairs or creating rapid reprompts.
To make your chatbot error-tolerant, try to anticipate common spelling mistakes first. Teach your chatbot these variations to improve its understanding.
- The process by which a speaker recognizes a speech error and repeats what has been said with some sort of correction. Ex. “Did you mean, ______?”
- Asking a short immediate question to give users another chance to respond. Ex. “Sorry, what was that?” vs. “Sorry, could you please provide me with your telephone number?”
Ready to implement these practices?
The hard stuff is over, now you're ready to build. Build your first chatbot with Engati. Click here to get started.