CONVERSATIONAL UI

Create a Powerful Conversational Experience for your clients

We partner with clients in various domains to create compelling conversational experiences. From simple informational chat or voice bots to more advanced bot platforms requiring ‘Natural Language Processing’ and complex integrations.

In searching for a more personalized and efficient way of interacting with their customers, a growing number of businesses are turning to Conversational Interfaces. These user interfaces can be anything from simple informational chat or voice bots to more advanced AI driven bot platforms requiring ‘Natural Language Processing’ and complex integrations.

Are you looking to maintain a human connection with your customers on all your platforms? Curious to discover the substantial gains a conversational user interface can provide to your business? Let’s get in touch. At Arinti, we help our customers build compelling conversational experiences in different domains like civil services, end-user support and HR-legal advice. We work with both large enterprises and smaller start-up companies in implementing innovative technological solutions to increase user satisfaction.

 

Developing bots on a modular framework

We develop our bots on an extensible modular platform that provides tools to build, test, deploy, and manage them all in one place. The platform allows us to build intelligent chatbots that provide speech, language understanding, question and answer, and that are tailored to our clients’ needs. It also gives us the opportunity to deploy our bots to multiple channels (websites, apps, social media, messaging tools).

Throughout the entire bot development process (design – build – test – publish – connect – evaluate), we act as a partner for our clients. Our engineers work in close collaboration with all stakeholders.

Step 1. Design

They key to designing a great chatbot is this: it needs to solve the problem your users face and it has to give them a great user experience. Otherwise, chances of your bot being used, will be very small. So, before designing or developing a bot, you need to make sure you are creating an experience that actually solves a business problem and does so in an easy, user-friendly experience. An experience your users will prefer over other alternatives, such as an app, website or phone call. Ensuring a great user experience should be your number one priority during the design phase of your bot.

In all of our projects, the design phase starts with a number of bot insights workshops. Through co-creation workshops with all stakeholders and ideation sessions with UX designers, we lay out the foundations of the perfect bot.

These are some of the key questions that are on the table during the design phase:

  • Does the bot solve the user’s problem in an easy and user-friendly manner?
  • Does the bot solve the user’s problem better/easier/faster than any of the alternative experiences (app, website, phone call, …)?
  • Does the bot run on the devices and platforms your target audience uses?
  • Is the bot discoverable?
  • Do your users intuitively know how to use it?

Note that none of these questions is linked to how smart the bot is. In most cases, it is unlikely that making your bot smarter will guarantee happy users or a better adoption of your platform. In reality, many bots have little advanced machine learning or natural language capabilities. And that’s perfectly fine! A bot may include those capabilities if they’re necessary to solve the problems that it’s designed to address. However, you should not assume any correlation between a bot’s intelligence and its user adoption.

Your bot can be great at conversations. It can have a vast vocabulary and maybe it even makes great jokes. However, unless it addresses the problems your users want to solve, these capabilities contribute very little to making your bot successful. In fact, some bots have no conversational capability at all. And in many cases, that’s perfectly fine.

You have to look at the business case for your bot to make sure you are designing your conversations and your bot experience in the right way.

Step 2. Develop your bot

In a second phase of the project, our developers will transform the design into a real, functioning solution. Developing a bot basically comes down to building a web service that allows you to implement a conversational interface into your application.

When building chat- and voicebots, there are a number of technical components that are crucial in making the solution work:

  • A database that contains the knowledge needed to answer a user’s questions
  • An advanced search algorithm
  • An NLP to add a language understanding model.
  • Adaptive cards or buttons
  • Speech-to-text and text-to-speech services, depending on the use case
  • Translation services
  • An object storage to make sure you can dive into conversation logs and data once your application generates traffic

The platform we us to build chatbots allows us to work with existing and well-established programming tools and environments, such as .Net (C#) or NodeJS and Visual Studio or Yeoman.

Step 3. Test

Bots are very similar to other complex tech applications. As they involve multiple integrations and many different parts that interact with each other, initial bugs can (and probably will) occur. That is why we take testing very seriously: before publishing your bot on your preferred channels, we test it over and over again in order to make sure everything works perfectly.

How do we test our bots?

  1. Through a desktop application that allows our developers to test and debug their creations, either locally or remotely.
    Using the application, you can already chat with your bot and inspect the messages it sends and receives. The application displays messages as they would appear in a web chat user interface. It also logs JSON requests and responses as you exchange messages with your bot.
  2. Through a test Web Chat channel that is configured for you when we create your bot.
    In this test channel, the bot will ask you to confirm the correctness of the given answer. That way, non-technical test users can actively contribute to the training of the bot.

Step 4. Publish

After testing, your bot can be published to the cloud or to a data center.

We can set up continuous deployment that allows us to develop your bot locally. This can be specifically useful if your bot is checked into a source control like GitHub or Visual Studio Team Services. As you check your changes back into your source repository, your changes will automatically be deployed to the cloud.

Step 5. Connect

Connecting your bot is about choosing the right channels.

Channels you can choose from, include many popular services, such as Skype, Facebook Messenger, Kik, Slack, Cortana, Microsoft Teams, Telegram, SMS, Twilio, and numerous others. Most of our customers currently use their bots in a dedicated Web Chat window on their website or intranet site.

Which channel to choose, is different for every case. While Facebook Messenger is often a great choice for an external chat bot. For instance, it can be used to give information about your company or to take the first basic questions from a job interview in order to streamline the recruitment process. Skype, on the other hand, may be more appropriate for an internal chat bot that can help your employees solve first line IT problems.

Step 6. Evaluate

Once your bot is live, you will want to know how it is performing.

We use the data collected in the bot portal and visualize it in a report to identify opportunities and improve the capabilities and performances of the bot.

Data that is gathered in the portal can be service-level data or instrumentation data, such as traffic, latency, and integrations. The data could also provide insights on a conversation-level by reporting on user, message, and channel data.

The collected data help us answer questions like these:

  • How many people use my chatbot?
  • How long does an average conversation last?
  • How many times was the chatbot unable/not confident enough to answer a question?
  • How many questions were about this specific intent/entity?