AIMultiple ResearchAIMultiple Research

Conversational UI: Best Practices & Case Studies in 2024

Cem Dilmegani
Updated on Mar 10
4 min read
Conversational UI: Best Practices & Case Studies in 2024Conversational UI: Best Practices & Case Studies in 2024

Over 50% of customers expect a business to be available 24/7. Conversational interfaces are an effective way for companies to have a round-the-clock online customer service and marketing, particularly for businesses with an international footprint.

In addition, employees are starting to leverage digital workers/assistants via conversational interfaces and delegate monotonous jobs to them.

Thus, conversational interfaces can improve consumer happiness while also increasing worker productivity. Explore conversational interfaces and its best practices to assist businesses:

What’s a conversational interface?

The conversational interface is an interface you can talk/write to in plain language. The aim is to provide a seamless user experience, as if you are talking to a human.

Most conversational interfaces today act as a stop-gap, answering basic questions, but unable to offer as much support as a live agent. However, with the latest advances in conversational AI and generative AI, conversational interfaces are becoming more capable.

How are conversational interfaces classified?

Conversational interfaces can be categorized into 2 broad categories: text-based assistants (also called chatbots), voice-based assistants (also called voice bots or voice assistants).

Image shows the evolution of UI.

Text-based assistants

Text is the most common kind of conversational interface between a human and a machine. The input consists of the words that individuals type. The chatbot presents users with an answer or clarification question based on the input.

Text-based conversational interfaces have begun to transform the workplace both via customer service bots and as digital workers. Digital workers are designed to automate monotonous and semi-technical operations to give staff more time to focus on tasks where human intelligence is required.

Voice assistants

Voice assistants such as Google’s assistant, Amazon’s Echo or more recently Rabbit R1, are being adapted by shopping sites and other websites.

While basic bots and text-based assistants can leverage images and video to convey their message, voice assistants have the downside of only relying on voice. Voice is sufficient for some use cases, such as re-ordering a frequently purchased item but it may not be a good interface for examining a new physical product like a dress or picking an item from a menu. For example, Dan Grover demonstrates that ordering a pizza takes 73 taps on a pure text interface and 16 taps from the Pizza Hut app which uses both text and images.

To learn more about conversational AI types you can read our In-Depth Guide to the 5 Types of Conversational AI article.

How can we classify the intelligence behind conversational UIs?

Bots typically fall under 3 categories.

Basic bots

Inputs for basic chatbots are rather limited. In most basic bots, users receive a list of commands to choose from. These can be used by applications with simple functionality or companies looking to experiment with a novel interface. These basic bots are going out of fashion as companies embrace text-based assistants.

Bots powered by traditional AI systems

These bots can answer more complex queries in their specific domain. Their responses are almost always correct if the context is clearly explained. They have been around since 2010s.

They are limited by their training data and may not be able to deal with edge cases.

Bots powered by generative AI systems / LLMs

These bots can engage in complex conversations in a wide variety of topics since they have been trained using a large volume of text. They are then finetuned to work as customer service assistants or sales bots etc.

They are prone to hallucinations and can make up non-existent policies (e.g. discounts or cancellation policies). Hallucinations can be costly and are among the most expensive conversational AI failures.

What are some case studies of successful conversational interface implementation?

We have an article covering top 25 chatbots but we highlighted an important example here:

KLM

Allowing customers to change seat or meal preferences, and get notified of flight delays, KLM’s chatbot is a useful conversational UI example for airlines.

Images represents an UI example of airline mobile app.

How to make great conversational experiences?

Some of the best practices you should be aware of:1

Building a bot is easy; building a useful one isn’t

Building a bot has gotten easier down the years thanks to open-source sharing of the underlying codes, but the problem is creating a useful one. It would take considerably long time to develop one due to the difficulty of integrating different data sources (i.e. CRM software or e-commerce platform) to achieve superior quality. The incomplete nature of conversational interface development also requires human supervision if the goal is developing a fully functioning system.

Narrow the domain

Making the chatbot as simple as possible should be the ultimate goal. This requires developing the conversational interfaces to be as simple as possible. The language the bot uses would shape the input provided by the user. So shaping the behavior of the user, by providing the right cues, would make the conversation flow smoothly.

Control for errors

Learning from mistakes is important, especially when collecting the right data and improving the interface to make for a seamless experience. Therefore, you should provide the right tools and feedback mechanism to correct errors and problems.

Structure your data collecting method

Structure the questions in such a way that it would be easier to analyze and provide insights. This can be implemented through multiple choice questions or yes/no type of questions. This would make it possible to analyze the data on the fly.

Make history matter

Leverage conversation history to modify the conversation flow. That way, your conversational interface would make the user feel as if she is chatting with an actual human being.

For more on conversational AI

To learn more about conversational AI, read:

Finally, if you are interested in adopting a conversational AI platform for your business, head over to our conversational AI hub, where we have data-driven lists of vendors for different purposes:

Find the Right Vendors
Access Cem's 2 decades of B2B tech experience as a tech consultant, enterprise leader, startup entrepreneur & industry analyst. Leverage insights informing top Fortune 500 every month.
Cem Dilmegani
Principal Analyst
Follow on

Cem Dilmegani
Principal Analyst

Cem has been the principal analyst at AIMultiple since 2017. AIMultiple informs hundreds of thousands of businesses (as per similarWeb) including 60% of Fortune 500 every month.

Cem's work has been cited by leading global publications including Business Insider, Forbes, Washington Post, global firms like Deloitte, HPE, NGOs like World Economic Forum and supranational organizations like European Commission. You can see more reputable companies and media that referenced AIMultiple.

Throughout his career, Cem served as a tech consultant, tech buyer and tech entrepreneur. He advised businesses on their enterprise software, automation, cloud, AI / ML and other technology related decisions at McKinsey & Company and Altman Solon for more than a decade. He also published a McKinsey report on digitalization.

He led technology strategy and procurement of a telco while reporting to the CEO. He has also led commercial growth of deep tech company Hypatos that reached a 7 digit annual recurring revenue and a 9 digit valuation from 0 within 2 years. Cem's work in Hypatos was covered by leading technology publications like TechCrunch and Business Insider.

Cem regularly speaks at international technology conferences. He graduated from Bogazici University as a computer engineer and holds an MBA from Columbia Business School.

To stay up-to-date on B2B tech & accelerate your enterprise:

Follow on

Next to Read

Comments

Your email address will not be published. All fields are required.

2 Comments
Caden Dahl
Feb 22, 2019 at 04:48

With conversation, it is amazing what we could do with it when it comes to AI. Now as you said here, there are multiple different platforms to where they are used. To me, I think that a voice assistant would be the most important as you could use it as a personal translator of some sort.

shunichi
Mar 16, 2018 at 18:09

Thank you so much Allie!! kisses

Related research