Chatbots are a incredibly hot topic in our industry right now. At this point we’ve all heard of bots, or even prototyped and launched a few of them to our clients and companies we work for.

Some people argue that chatbots are the new websites and that they will kill 99% of the apps out there — predicting that conversational interfaces will soon replace the pixel-centric design patterns we’ve been using for decades in our work.




But then, this:






And this:






Well, and this:






Why do chatbots fail?

The idea of automating and scaling one-to-one conversations using technology appeals to lots of brands and services out there. In this process, designers play an important role at defining how each conversation is scripted and the behaviors users can expect when interacting with bots.

But, despite the best of our intentions, sometimes chatbots fail to deliver user experiences that are as seamless, delightful and efficient as we envisioned them to be.


What is going on?







Here are a few common reasons for chatbots to fail:




Artificial Intelligence (AI)
is still not that accessible

The vast majority of chatbots aren't actually intelligent. They are built based on a decision-tree logic, where the response given by the bot depends on specific keywords identified in the user's input. 

IF user's input contains 'shop' or 'buy';
THEN send message with product list

What that means is that decision-tree types of bots are as intelligent as the capacity (and thoroughness, and patience) of the designer/programmer who created it to anticipate all potential user use cases and inputs.

Bots with linguistic and natural language learning capabilities are still quite rare.

I hope it's sunny where you are

I hope it's sunny where you are


Use cases are not that strong

Here's something that happens with every new technology that is put out in the world: designers and developers get really excited about it.

What we are seeing now is a gold rush of companies trying to be the first in their category to successfully deploy a bot. In that process, we will see a plethora of bots that are solving for irrelevant use cases, or that offer really poor experiences.

It’s a natural part of the cycle: our industry needs to learn from its failures before it is able to deploy bots that are truly relevant and smart.

A bot for bad dad jokes, or simply a 'dadbot'

A bot for bad dad jokes, or simply a 'dadbot'


Some bots lack transparency

The most successful bots out there make one thing clear from the very beginning of the experience: that the user is chatting with a robot, not with another human. Setting up the right expectations upfront will make users more forgiving about certain mistakes the bot might make.

You certainly want your bot to feel as human as possible, but lying to your users and pretending to be something it is not can lead to irreversible loss of trust.

Our very own  pretending it is really listening to what users are saying

Our very own pretending it is really listening
to what users are saying


They don't understand context

Humans are really good at conversations. We understand sarcasm, we can read between the lines, and we are constantly leveraging contextual information when we give someone a response.

When I'm arranging dinner plans with someone over the phone and I ask if I should bring my umbrella, the person knows where we are going, what time of the day we are meeting, and whether that is an indoor or outdoor venue. They even know how cautious of a person I am – and take all that into consideration when giving me an answer.

Bots do not. 

Except in cases where bots are powered by natural language processing technology, they can't hold contextual information for longer than a few chat bubbles, and will end up losing track of what the user was saying before they posed the question.

Dory , the bot

Dory, the bot


They don't communicate with existing business systems

Another common temptation when building a chatbot is trying to recreate functionality from scratch.

Let's say you are creating a bot to book appointments in a spa. If your chatbot does not communicate with the spa's existing appointment management system, that means extra work for the business owner to handle requests coming through this new channel – and ultimately lack of consistency for the user.

Bots are part of a larger ecosystem, formed by multiple touch points between customers and brands. Creating a chatbot in a silo can be pretty harmful for both businesses and customers.


They try to handle
too many things at once

 Designers and developers tend to get excited about all the tasks a bot can help with, but forget to narrow its area of focus. Don't try to address problems that go beyond your scope.

Bots that do one thing well are more helpful that bots that do many things poorly. 

you can actually replace "bots" in the sentence above with "apps", "sites", and even "people"













They lack proper
human escalation protocols

When technology fails, users still want to be able to rely on human beings to help them solve their problems. Still, very few chatbots have an escalation workflow in place to let a human take over the conversation when the bot is unable to help.

The result?

Bots that leave users hanging – sometimes even more frustrated than when they started the conversation with the brand.





What that means for us, designers

Designing for conversational interfaces represents a big shift in the way we are used to thinking about interaction. Chatbots have less signifiers and affordances than websites and apps – which means words have to work harder to deliver clarity, cohesion and utility for the user. It is a change of paradigm that requires designers to re-wire their brain, their deliverables and their design process to create successful bot experiences.




So where do I start?

Well, here's where your friends at come in.

Over the last few months we have prepared a special collection of articles to help you in the journey of building better, more relevant and delightful chatbot experiences.

The collection includes tutorials, references, examples, and stories from the field written by a team of experts in our industry who have learned from their own successes and failures.


Let's get started with chatbots >

Hope you enjoy,



Share the love ❤