Photo by Samuel Charron on Unsplash

Chatbots, Designed Paths & Desired Paths

And How HumanFirst Is Pioneering Intent Driven Design & Development

Introduction

Conversational designers have training and expertise in crafting engaging conversations. Generally conversations are crafted around products and services. Hence a big part of the process is to improve the conversations by focusing & improving the design.

When contextual awareness is enhanced via accurate intent detection and contextual back-end lookups, the conversational surface exposed to the user can be limited & directed.

For conversational interfaces, a big cause of missed intents from user utterances are new product and new services. The problem here is customers want to chat to your chatbot based on advertising and marketing, but the intents have not been updated.

Also, the “customer speak” is not yet known for the product or service.

Clustering user utterances based on similarity affords the opportunity to find new clusters without intents assigned to it. Setting the time period, helps to pinpoint aberrations and changes in customer input.

Designed & Desire

The designed path is the conversation designers and developers envision users would want to have with their chatbot.

Often, in reality the user’s desired path can be something quite different.

The Designed Path opposed to the user’s Desired Path illustrated.

I was once involved in launching a chatbot for a pizza franchise, we discovered two intents which were not catered for. These were job seekers and requests for sponsorships. South Africa has an unemployment rate of 32%, and a youth unemployment rate of 58%.

These figures are the official numbers; but in reality it is much-much higher.

Hence this regional problem transpired in the conversations and it had to be catered for. This is a good example where the designed path was not aligned to the user’s desired path.

The design is aligned with the organizations objectives and with what is deemed good conversational design. The conversation the users want to have is often in stark contrast.

Alignment of what the user says, and what conversations are designed.

Looking at the Venn Diagram above, there is what he users want to talk about. Then there is what user’s are saying. And the overlap is where the magic happens.

Looking at the Venn Diagram above, it is clear that these two circles need to overlap 100%. The overlap needs to increase over time rather than decrease in size.

The best way of aligning the two circles is astute and accurate user utterance and conversation clustering with appropriate granularity. Hence the importance of Intent Driven Design & Development.

Intent Driven Design & Development

Journeys or dialog paths need to be seen as disposable and not deemed precious by designers. If the dialog or journey does not serve the conversation the user wants, it needs to be disposed of. Or adjusted to suite the desired path.

HumanFirst is pioneering Intent Driven Design & Development. This is where user utterances and conversations are used (from any CRM system not just Conversational Interfaces) to accurately detect utterance clusters.

These utterance clusters or groupings constitute intents. These intents are a clear indication on what users want to have a conversation on.

As the graph above shows, it is perceived with mobile apps that efforts is primarily upfront while with chatbots the real work starts after launch.

Looking at the graph above, and comparing time and effort in developing a mobile app or chatbot… There is an initial escalated effort to get the mobile out into production. However, over time the effort diminishes.

The inverse is true about conversational interfaces. The effort is initially low to release the interface. The real effort starts when the interface is exposed to the rigors of real-world customer conversations. And later in the process the effort increases to address bad customer experiences.

This is where HumanFirst comes into its own, extracting user intents from clustering user utterances or conversations. And flattening the curve of determining relevant journeys becomes easy. All with a no-code Machine Learning approach.

User utterances can be assigned to intents, or clustered automatically without any intents defined prior.

The ideal is to cluster user utterances and conversations gleaned from other mediums like live agent chat, web searches etc. right at the onset of the process. Hence mimicking the mobile app development effort trajectory. The advantage of this is negating the initial poor user experience.

Complex Intents

When presented with a menu of very applicable options, less than 5% of users opt to selected one of the menus, in some studies. From the total amount of users, 95% opt to rather express their desire in words using a sentence or two.

Users opt to rather enter unstructured data.

Virtually all users opt to not structure their input according to a graphic menu structure, but merely type their request in natural language. Obviously this is what a conversational agent is designed for.

Also, I suspect users conversing in a second or third language are having trouble to categorize their request according to the menu items available.

This scenario is exacerbated when technology and specific products and services are involved. Hence users would rather defer the structuring of the input to the chatbot.

With the user describing their problem or query, compound intents and entities are introduced, which need to be handled by the chatbot.

Creating structure from unstructured conversational data can demand complex intents. Demanding nested intents. Flexibility of merging or creating nested intents by dragging & dropping intents is real important.

Creating structure from unstructured conversational data can demand complex intents. Demanding nested intents. Flexibility of merging or creating nested intents by dragging & dropping intents is real important.

As user input changes over time, defined intents and entities need to be flexible enough to be adjusted accordingly.

Here is an example where user utterances are clustered automatically without any prior intents defined.

Here is an example where user utterances are clustered automatically without any prior intents defined.

Disambiguation

When users enter longer sentences, or describe their problem as a non native English speaker, ambiguity is introduced in many instances.

Examples of Ambiguity on the left. And how IBM Watson Assistant is presenting a disambiguation menu to the user.

It needs to be noted that there is ambiguity which is easy for humans to disambiguate, but very hard for machines. The first example above being a case in point.

Then, there is legitimate ambiguity which is impossible to interpret (the subsequent binocular example). Here disambiguation needs to be employed.

The chat interface above left demonstrates how an ambiguous entry can be disambiguated with a short menu.

Domain & Irrelevance

Conversational Agents will virtually always be domain specific. All user entries will either fall within the domain or outside.

The domain of implementation needs to communicated to the user, rather than try and match the closest intent.

Some chatbot environments give the option to mark user input during training as irrelevant. This does not mean that the domain must not be adjusted as the users’ desire is detected and the domain needs to migrate to cover new intents.

Some chatbot environments give the option to mark user input during training as irrelevant. This does not mean that the domain must not be adjusted as the users’ desire is detected and the domain needs to migrate to cover new intents.

Conclusion

Intent Driven Design and Development allows for the most appropriate journeys to be designed and developed.

Time spent on design and improvement is justified with the knowledge that the journey is important to the user.

Journeys are mostly invoked by a condition, invariably involving at least one intent. Intent Driven Design and Development allows for the right journeys to be presented to users.

--

--

--

NLP/NLU, Chatbots, Voice, Conversational UI/UX, CX Designer, Developer, Ubiquitous User Interfaces. www.cobusgreyling.me

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Redesigning Sync

Brainstorm

Case Study: To increase blood donorship following the COVID Pandemic

A wooden hand holding an iphone displaying an interactive map with blood donation clinics.

Ironhack PreCourse — Challenge 2: Wireframing

Immunisation Consent Discovery: Part 2

A computer monitor covered in research goal post-it notes. Some marked “S” for school staff, some marked “P” for parents

Prototyping

Should you become a UX designer and do you have what it takes?

An illustration of a phone showing a dating profile for the job of UX design

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Cobus Greyling

Cobus Greyling

NLP/NLU, Chatbots, Voice, Conversational UI/UX, CX Designer, Developer, Ubiquitous User Interfaces. www.cobusgreyling.me

More from Medium

Use Customer Utterances To Create Chatbot Intents & Intent Examples

Watson Assistant: Connecting Customer Care to IBM Cloud Functions APIs

How We Build a Conversational AI Chatbot at MoMo

Barge-in for Voice Assistants and Voice IVRs