The leaders of virtually every tech giant are on a mission to convince us that chatbots are the future. “You should message businesses just as you would your friends,” said Mark Zuckerberg when he launched Facebook Messenger Platform for bots, and they’re not wrong because conversation can be more immersive than ever before!
But is chatting actually good for bots? Early user reviews of chatbots suggest not. Gizmodo writer Darren Orf describes Facebook’s bot experiences as “frustrating and useless”, comparing using them to trying to talk politics with a toddler. A pretty fair assessment! His criticism isn’t without merit either. Early adopter users Report missing content messages or getting stuck in long loops easily. But these are hardly surprising flaws given how novel this technology still feels.
To Chat Or Not To Chat?
Designers who are new to conversational interfaces often assume chatbots must “chat”. They also underestimate the extraordinary writing skills and technical investment needed for excellent UX, which requires continuous iteration in order not just implement but perfect over time as well.
Chatbots have come so far since their inception five years ago! In that short amount of time we’ve seen them transform from simple automated customer service into interactive agents capable of understanding natural speech patterns with eloquent responses based on context-sensitive logic.
When it comes to conversation, there are two sides of the story. One side advocates for more open-ended chats while others suggest that bots should prioritise consistent conversations. Keep users engaged by giving them what they want quickly in a one-size fits. All approach so you don’t lose out on potential sales or engagement opportunities with your customer base. They navigate around different features within the app which could lead people down dead end paths if not done properly.
Chatbots are becoming more and more popular, but many people haven’t experienced the benefits of this technology. A chatbot can either be very helpful or frustrating depending on how it’s implemented. Which makes choosing an appropriate level for chatbot important!
The Case For Chat
The world’s most popular chatbot, Mitsuku, has twice won the Loebner Prize. With over 5 million users and 150 million total interactions processed to date it is clear that 80% of all conversations come back again with more messages from this interesting conversational system.
“The longest a user has chatted with Mitsuku is 9 hours in one day, and it’s all thanks to his extraordinary conversational abilities.”
Mitsuku doesn’t help you find make-up products or buy flowers. His sole purpose of providing entertainment and companionship through chats on Messenger! You won’t be surprised when we say that Worswick thinks “chatbots should be about the chat”
Building a conversational chatbot that is not awful can be hard. Mitsuku nearly gave up many times when users called her “stupid” and she had to answer their questions in an unsatisfactory way time after time again while Worswick programmed with database of objects like chairs, trees or cinemas along with their relationships attributes which made the answers more correct for people who were asking already knew how those things worked.
Mitsuku has really come into her own since the implementation of artificial intelligence. She can now give sensible answers to strange user questions, such as “Is a snail slower than a train?”
What would you say if we told you that your favourite cat is on the internet? Poncho, an incredible Messenger bot with a personality and voice just like yours! The creators gave him the ability to do small talk for when boredom sets in during those rainy days. But don’t worry because he also recognizes other cats from around town– so all will be entertained even without any nonsense talks or jokes!”
When You Should Add Conversation To Delight Users
If you’re ready to take on the challenge of making a bot conversational, here are reasons why it will set your chatbot apart from competitors and delight users.
If you need to differentiate from competition
Poncho is a chatty weather cat with an interesting personality that sets him apart from boring and routine weather apps. The bots launch at a more rapid pace than mobile devices due to lower technical barriers of entry, so it’s important for winners in this category to stand out by creating superior conversational UX.
Just like the weather, public transit apps are soulless and boring. We use them out of necessity but often find ourselves feeling disengaged from our surroundings with each tap on an icon or tapping away at numbers to see when a bus will arrive next in this modern age where everyone strives for convenience over creativity! Enter Bus Uncle: A bot who can tell you anything about buses anywhere across Singapore (in his quirky broken English) while suggesting all sorts of things that might help pass time waiting such as watching videos.
The Bus Uncle app is like having your own personal guide. You can find all of the bus stops in a given area, filter by type and time schedule available – even input parameters such as how many people are travelling with you!
Logging onto the map will show which routes connect those hubs so that it’s easy for anyone who is unfamiliar (or just not tech savvy) to figure out where they need to go next without getting lost along the route path before being able to get there themselves.
The Bot Uncle may be all fun and games at first, but he stays in character by forcing you to do maths problems from time-to-time.
If you need to handle edge conditions
Conversational shopping bots have the potential to be one of the most compelling tools in e-commerce, but they must overcome many challenges before achieving that goal. For example: Bots are often not smart enough or equipped with all data needed for product recommendations when users use unfamiliar words or switch contexts suddenly without warning; Yet another problem is how these systems handle edge user input which may lead them going off topic.
ShopBot from eBay is a bot that aims to avoid common e-commerce user experience failures by combining limited option menus. The ability to handle unexpected input. While many shopping bots hem users in, limiting their choices and putting pressure on them at all times. Shopbot can adapt when switching from browsing jeans’ selection to browsing shirts. Providing more flexibility for customers who want both types but don’t have time or patience enough just yet make up their mind about what they need before going into store!
Chatbots are not yet perfect for e-commerce. While they do an excellent job at handling simple tasks, such as asking about the weather or providing information on what’s playing at any given time in theatre listings -chatbot performance begins to decline when you need them most: trying to purchase something from your favourite online retailer!
Superior conversational experiences require powerful technologies that process natural language and keep track of shoppers’ contexts while also anticipating diverse needs accurately so we know how much things cost across different locations without having to search every single page again.
eBay has announced that they plan on using artificial intelligence in order to provide shoppers with a more personalised experience. The company’s Chief Product Officer, RJ Pittman says “Shoppers have complex needs which are often not fully met by traditional search engines.” With the help of AI he believes this will soon change because machine learning abilities allow for contextual understanding and predictive modelling so users get what they want without having too many worries.
If you can be relevant and timely
The race for president in America is one of the most watched and talked-about elections globally. Inside Amazon’s Echo HAL scraps through information on who won each debate, giving you live updates about polling data as well!
Users took advantage of the timely features to ask Alexa open-ended questions such as “How old is Trump?” and “What’s going on in politics today!,” but more often than not, users were asking “Alexa who are you voting for?” Also ironic was how many people wanted answers about themselves compared with other topics like climate change.
When it comes to the latest topics, bots that feel like they really know what’s going on will be better at engaging their users than those who regurgitate facts or offer links.
If you can humanize a brand
Chats are intimate interactions where you can feel connected to others. Chatting with a brand often feels odd because of the lack in conversation skills and personality that goes into it, but strong conversational abilities will help break down barriers for users who want more from their customer service experience.
Maintaining a consistent and compelling brand voice in chatbots can be tough. PullString, an artificial intelligence (AI) platform founded by Oren Jacob who was the CTO for Pixar Animation Studios is employing more than just writers- it’s also using speech recognition software which will allow them to produce personalised responses depending on what people want from you!
They created a demo chatbot, Jessie Humani to demonstrate how they can produce an artificial intelligence-powered virtual assistant who seems like your average millennial. Powered by over 3 500 lines of carefully selected dialog and designed with messiness in mind – it’s not hard for the bot impression you get when dealing with this character!
The Case Against Chat
The word “chatbot” sets the wrong user expectation that bots should have human-level conversational abilities. The hard reality is that natural language processing and artificial intelligence still have much progress to make before they will be able to impress you with their gift of gab.
Ted Livingston, CEO of Kik—a messaging platform that has a thriving bot store–has found success in educating users on how to use bots. He shared with an audience at the recent Bot conference: “The biggest misconception is that chatbots need to be about ‘chat’. What we discovered was once you remove all suggested responses and only leave room for input field + blinking cursor; people have trouble using them because there’s nothing interactive.”
Kik is an app that started out as a simple chat platform. They created their own bot system for people to message back and forth. But soon found it difficult with all the open-ended conversations happening on Facebook Messenger bots these days because they are not bound by the same rules or limitations.
For example, when Sephora’s Kik bot asks what type of beauty products a user would like to see, the bot follows the question with a menu of suggested responses to choose from. A user has to go out of her way to hit “Tap a message” in order to type normally.
When You Should Restrict Chat For Better UX
Restricting conversation to a few key responses. The best way for designers of chatbots to provide a superior user experience. To accomplish this, there are several common situations where letting users type freeform conversational text complicates development and decreases your bot’s usability.
If user errors lead to failed transactions
flower delivery bot for facebook messenger. Originally gave users three options: “Today”, “tomorrow” and a date selection option that allowed them to type in their own preferred times. The third choice let people choose from one of these free-forms. Which often led to confusion with errors or abandoned transactions!
By removing the third option for users to type in arbitrary dates, Flowers actually increased their customers’ satisfaction. Restricting conversation helped them focus on what mattered most. Urgent flower delivery!
If your competitive advantage is simplicity
Chatbots should give users the key advantage of completing tasks with fewer taps and context switches than regular mobile apps. This simplicity is enabled by allowing open-ended chats in contrast to text input on a particular topic. Which can be limiting for some use cases but an excellent choice. When you’re looking for versatility or have many clients that need different things.
The meditation bot Peaceful Habit for Amazon Echo and Facebook Messenger is an excellent way to help you. It can be quicker than apps, since all the information needed will automatically pop up on screen during use!
On the Amazon Echo, a user can start a 5, 10, or 20 minute meditation completely hands-free, with voice alone. On Facebook Messenger, the bot sends a daily reminder with limited user options so only a single tap is required.
With a messaging Chatbot. Users can create their very own meditation session with just the choice of duration and start time. Themed according to preferences or seasons, these bots make it easy for you. As well as others who might need help calming down!
If you cannot easily handle unbounded input
Many user requests appear simple on the surface but are extremely complex to handle. In open-ended conversational interfaces due to vocabulary, grammatical structures and cultural norms. For example a request for scheduling an appointment by asking following questions:
When’s Bob’s next open time slot?
Let me know the next three times Bob can chat.
Is Bob available at 4 PM PST today?
Now, it turns out that the complexity of handling seemingly simple meeting requests requires powerful artificial intelligence and human expertise. Several well-funded companies have emerged just to solve these problems. With specialized technology like natural language processing or reasoning systems powered by machine learning algorithms. Which can be used in scheduling meetings efficiently!
However, this can be less of an issue when the bot offers to limit conversations. For example, if your customer wants recommendations for local restaurants that are within walking distance or offer vegetarian options – nice! The AI will know what they’re looking for and help them to find it much faster way.
By contrast, a similar bot called OrderNow finds local restaurants that deliver and offers up a limited menu of cuisines to choose from.
Some people think that complex artificial intelligence, machine learning is required to create great user experiences for chatbots. As Ted Livingston warns “AI isn’t the killer app of bots; it holds most back.”
The reality? Chatbot software should focus on delivering an excellent experience with one thing well-constructed about their functionality. Doing what they’re made for without any fuss and bother!
Chatbots are the future of customer service, but you don’t want to be seen as a chatbot. People crave human connection and conversation is what they’re looking for in their interactions with bots. If people can tell that it’s just a bot on the other end of an online chat interface. Company could lose potential customers who would rather deal with someone else or simply not make any purchase at all. This is why companies must take care when deciding how much conversational content should go into their chats. Since too little will frustrate users while overloading them with information may scare them away. There are ways around these problems by having one employee monitor conversations so that they know when to step in and offer help.
Other Resources: https://www.blogpostdaily.com/category/computers-and-technology/