UX AI Transformation: The Rise of Conversational Interfaces
This post covers one of four transformations and is part of a larger blog series titled 'From Hype to Reality: A Practical Guide to AI-driven UX in SaaS'. You can find the first post of the series here.
Conversational Interfaces in SaaS: AI's Influence on User Interaction
Technology's become such a big part of our lives that we want to chat with our devices like we do with people. That's where conversational interfaces come in. Instead of following rigid steps or tapping on buttons, we can now just talk to our software like we would with a friend. This shift, powered by advancements in Natural Language Processing (NLP) and AI, has led to a surge in conversational interactions.
Chatbots, the helpful ones, are popping up everywhere— salesforce, microsoft, IBM, intercom, zendesk, hubspot, freshworks,, slack, zoho, zapier, ada, tars, manychat, overloop and many others are integrating them into their platforms. The rate of worldwide retail spending on chatbots is set to soar from $12 billion in 2023 to $72 billion by 2028 according to emarketer
They're not just for answering simple questions anymore; now we have Co-Pilots, more advanced chatbots that can handle complex tasks. First seen only in the market leaders: Salesforce, IBM, microsoft, amazon, now increasingly being developed across a diverse spectrum of SaaS, from customer service to project management, including clare, dixa, chatfield, bitrix24
Shaping User Experiences: The Influence of Conversational Interfaces
Users love the natural feel of conversational interfaces. It's like having a real conversation, but with your software. Research shows that people find these interactions more engaging and personal. They can get exactly what they need without navigating through menus and buttons. Plus, it's a game-changer for accessibility, making digital experiences easier for everyone. This study found that 70% of users chose ChatGPT-powered conversational interfaces over traditional techniques, citing convenience, efficiency, and personalization.
But it's not all sunshine and rainbows. Sometimes, these interactions can leave users feeling frustrated or confused, or feelings such as distrust, intrusion, inconvenience, as indicated by this literature review. Open-ended dialogue boxes can be vague, making it hard for users to know what to say. And not everyone finds it easy to express their needs in words. Chat GPT, the tool that propelled the conversational experience, was never meant to be a consumer product (It was a demo to internally test out the LLM versions) and is a good example of a tool, that whilst is very useful, has many usability issues
Crafting Conversational Experiences: Insights for UX Designers
As cool as conversational interfaces are, they're not always the best solution for every situation. Designers need to be careful not to jump on the chatbot bandwagon just because it's trendy.
It's crucial to test these solutions rigorously and learn from existing research. NNG research often writes about the usability issues with conversational experiences and Why Chatbots Are Not the Future covers some of the main issues.
Companies like Perplexity are paving the way with user-first AI products that prioritize simplicity and speed. Designers should take notes and focus on providing answers rather than creating complex conversational experiences. Henry Modisett,Head of Design at Perplexity AI, talks through his approach to Perplexities design here. By staying grounded in user needs and usability principles, designers can navigate the world of conversational interfaces more effectively.