our blog

The Role of Research in Designing AI That Works for People

The Role of Research in Designing AI That Works for People

Designing an AI-powered interface is about ensuring that smart features enhance the user experience. It’s about creating AI-driven platforms that are intuitive, accessible and engaging, with the goal of improving the user journey. This means understanding the problems you’re trying to solve - whether it’s accessibility, engagement or usability. The primary aim is to make sure AI enriches the experience without overwhelming the user. Defining these challenges early on shapes the design direction, ensuring AI becomes a tool that empowers, not complicates.

Take, for example, our work with Diversity Economics, a B2B platform helping organisations track and refine their diversity, equity and inclusion (DEI) goals. One of the main challenges businesses face is the lack of structured benchmarking data, making it hard to assess DEI performance and implement meaningful change. In designing this platform, we focused on overcoming barriers in data accessibility, engagement and usability to ensure a seamless experience for companies seeking to improve their DEI strategies. Our approach involved researching Gen AI tools to identify their strengths and weaknesses and incorporating the best aspects into the platform’s design to make it both user-friendly and informative.

Once the problem is clear, it’s essential to evaluate existing solutions. By testing user interactions with AI driven features and identifying pain points, we can refine our designs to improve usability. For the DEI benchmarking platform, we designed an interface that allows businesses to compare DEI performance across industries, interact with a conversational AI assistant for guidance on best practices and compliance, and upload internal documents for AI powered analysis of language bias. This process helped us to pinpoint features that truly enhanced the experience and align the platform’s functionality with real-world user needs.

AI needs to collaborate with the user, not compete with them. For this project, we made sure the platform’s AI was intuitive and flexible, enabling users to easily understand what the AI was doing and how to use it. Transparency was key - users could access explanations for AI driven decisions, and they had the option to adjust AI suggestions to better suit their needs. This flexibility ensured the platform didn’t feel like a “black box” but a useful, adaptive tool.

The interface itself was designed to be simple and predictable. For the DEI platform, we built a seamless, interactive experience where businesses could navigate data visualisations, access dynamic reports and engage in guided learning sessions to support their DEI strategies. Performance was crucial - AI should enhance the speed and efficiency of tasks, never slow things down. By ensuring the platform ran smoothly and responded quickly to user inputs, we provided businesses with realtime insights into their DEI efforts. Creating a successful AI interface is a continuous process. For the DEI platform, this meant ongoing testing and iteration based on user feedback. 

As AI evolves, so should the platform, so we made sure the design was flexible enough to incorporate new advancements and features. The balance between automation and human control was essential, allowing users to override AI suggestions when necessary, giving them full control over the process.

At the heart of good design is research. By understanding user needs, evaluating existing tools and continuously iterating on feedback, we were able to create a platform that not only helped businesses with DEI benchmarking but made the process more accessible, engaging and actionable. The DEI platform’s success shows how AI can be a powerful ally in transforming user experiences, but only when thoughtfully integrated with a focus on usability and flexibility.

Good design starts with understanding the problem and building from there. Whether for DEI benchmarking or other uses, a well designed AI interface ensures the technology works for the user - helping them achieve their goals with ease.

spread the word, spread the word, spread the word, spread the word,
spread the word, spread the word, spread the word, spread the word,
Abstract illustration showing multimodal AI interfaces connecting voice, chat and automated systems across a digital product workflow, representing unified interaction design
AI

Multimodal AI Interface Design: Connecting Voice, Chat And Automation In Digital Products

Abstract illustration of AI-driven software interfaces showing uncertainty in outputs and decision-making across digital products.
AI

How AI Interfaces Are Changing Certainty In Software Design

Abstract illustration of AI-driven product interfaces showing connected systems of interaction across chat, workflows and adaptive outputs instead of traditional screen-based journeys.
AI

AI Is Changing How Products Actually Work

Workflow diagram illustrating AI agents producing outputs with human oversight and structured intervention points
AI

When AI Agents Get It Wrong

Workflow diagram showing multiple AI agents being monitored with human oversight
AI

Running AI Agents Reliably in Production

Multimodal AI Interface Design: Connecting Voice, Chat And Automation In Digital Products

Abstract illustration showing multimodal AI interfaces connecting voice, chat and automated systems across a digital product workflow, representing unified interaction design
AI

Multimodal AI Interface Design: Connecting Voice, Chat And Automation In Digital Products

How AI Interfaces Are Changing Certainty In Software Design

Abstract illustration of AI-driven software interfaces showing uncertainty in outputs and decision-making across digital products.
AI

How AI Interfaces Are Changing Certainty In Software Design

AI Is Changing How Products Actually Work

Abstract illustration of AI-driven product interfaces showing connected systems of interaction across chat, workflows and adaptive outputs instead of traditional screen-based journeys.
AI

AI Is Changing How Products Actually Work

When AI Agents Get It Wrong

Workflow diagram illustrating AI agents producing outputs with human oversight and structured intervention points
AI

When AI Agents Get It Wrong

Running AI Agents Reliably in Production

Workflow diagram showing multiple AI agents being monitored with human oversight
AI

Running AI Agents Reliably in Production

Multimodal AI Interface Design: Connecting Voice, Chat And Automation In Digital Products

Abstract illustration showing multimodal AI interfaces connecting voice, chat and automated systems across a digital product workflow, representing unified interaction design

How AI Interfaces Are Changing Certainty In Software Design

Abstract illustration of AI-driven software interfaces showing uncertainty in outputs and decision-making across digital products.

AI Is Changing How Products Actually Work

Abstract illustration of AI-driven product interfaces showing connected systems of interaction across chat, workflows and adaptive outputs instead of traditional screen-based journeys.

When AI Agents Get It Wrong

Workflow diagram illustrating AI agents producing outputs with human oversight and structured intervention points

Running AI Agents Reliably in Production

Workflow diagram showing multiple AI agents being monitored with human oversight