How to Maximize Understanding with AI for Innovation

Understanding people's experiences, behaviours, and perceptions has become more important—and more complex—than ever in today’s world. As organizations are flooded with vast amounts of customer data from an ever-growing number of touchpoints, making sense of it all has become increasingly difficult. Qualitative research remains the gold standard for gathering rich, non-numerical, meaningful insights, but now, AI is stepping in to streamline and enhance certain parts of this process

However, AI alone cannot (yet?) fully grasp context and nuance. While it can efficiently gather and categorize vast data, AI struggles to replicate the depth of human interpretation needed for making strategic decisions. Here’s how we can leverage both to drive meaningful insights that guide innovation and success.

The Role of Qualitative Research

Professionals such as UX Researchers, Designers and Product / Customer discovery experts are often trained and specialize in methodologies like in-depth interviews and ethnography to extract deep insights. Their key contributions include:

• Methodologies: Skilled in various qualitative techniques, they engage with participants to collect rich, descriptive data.
• Data Analysis: They identify patterns and themes using methods like thematic analysis and narrative analysis.
• Interpretation: They interpret findings within context, considering cultural and social factors that influence responses.
• Reporting: They present insights in compelling ways, often using quotes, narratives and other tools to illustrate key points.

Practical Applications

In practice, qualitative research serves as the foundation for understanding complex human behaviours and decision-making processes. By exploring the 'why' behind actions and preferences, it enables businesses and organizations to uncover deep insights that drive innovation, improve experiences, and inform strategic decisions across various areas—from product development to marketing, sales and beyond. These insights are critical for aligning business strategies with real-world needs, ensuring long-term success and adaptability in a competitive market.

• Innovation & Strategic Foresight: Qualitative research provides valuable insights that guide innovation and inform strategic decisions, enabling organizations to adapt and thrive in a changing market.
• Product Development & UX Research: Helps identify user needs and pain points, informing product direction and improvements. Observing user interactions helps identify issues, gaps, and opportunities, informing design changes.
• Marketing Strategy: Understanding customer motivations aids in crafting targeted messaging.
• Branding: Insights from qualitative research guide brand positioning and identity development.
• Customer Experience: Gathering feedback on customer experiences allows businesses to identify areas for enhancement.
• Employee Engagement: Understanding employee perceptions can inform practices and shape organizational culture.
• Competitive Analysis: Insights help refine strategies based on customer perceptions and competitor positioning.
• Trend Analysis: Exploring emerging trends enables businesses to stay ahead of market shifts.
• Sales Discovery: Sales professionals can utilize qualitative techniques to understand a prospect's needs.

How AI Can Enhance These Processes

AI can streamline certain aspects such as:

• Data Collection: Automate routine tasks to save time on manual and administrative work.
• Data Analysis: Use AI to analyze large volumes of qualitative data, efficiently identifying themes, sentiment, and key phrases.
• Reporting: AI can assist in generating preliminary summaries and insights, making it easier to share findings with stakeholders.

The Human Element in Qualitative Research

Despite these advancements, certain aspects of qualitative research still require the human touch.

Some conversations need to be between humans. Tones need to be heard, behaviours noted, facial expressions seen, and energy felt.

• Contextual Understanding: AI lacks the ability to fully grasp the nuances of human experiences, cultural subtleties, and emotions, which are vital for accurate interpretation.
• Building Relationships: Conducting interviews demands empathy and rapport-building, skills that AI simply cannot replicate. Additionally, it’s crucial to build relationships internally with teams and externally with customers and partners. For example, when I was leading UX Research & Product Insights at a B2B SaaS company, I built relationships with Customer Support, Customer Success, Sales, Marketing, and more. Going beyond the data from support tickets was imperative to understand nuances by talking to customer support reps. This also gave me insight into the end-to-end customer journey across the organization, which I ultimately mapped.
• Interpretation: While AI can identify patterns, it cannot understand the significance of findings in the same nuanced way that a human can. Qualitative research often relies on understanding the 'why' behind responses, necessitating human insight.
• Ethical Considerations: Navigating ethical dilemmas in qualitative research requires human judgment and sensitivity, especially when dealing with vulnerable populations.

But... why not just give everyone access to all the feedback?

Delivering direct, unfiltered feedback can have clear benefits, like providing raw insights without the distortion or bias that may occur when feedback passes through multiple teams.

However, there are several cons and potential dangers associated with this approach:

• Over-reliance on Feedback from Vocal Customers: Unfiltered insights often come from the loudest voices, which might not always represent the majority of customers. If teams focus too much on raw feedback from a few vocal customers, they risk missing out on more balanced perspectives. Over-reliance on feedback from a small subset of customers can skew product development or customer service priorities, potentially alienating the broader customer base whose feedback is less prominent.
• Risk of Misinterpretation: Direct feedback may lack the context needed for teams to interpret it accurately. Customers might provide vague or unclear comments that, without proper filtering or explanation, can be misinterpreted, leading to wrong conclusions and actions. Misinterpreted feedback can result in misguided changes to product, service, or customer strategies, leading to poor decisions that could negatively impact business outcomes.
• Overwhelming Volume of Information: Receiving raw, unfiltered feedback can result in overwhelming amounts of data, especially for teams already stretched thin. Critical insights might be missed.
• Lack of Prioritization: Unfiltered feedback doesn’t prioritize the most important issues. All data is delivered with equal weight, meaning less critical feedback might distract teams from focusing on more urgent problems. Teams may spend time addressing low-priority concerns while neglecting higher-impact issues. This can lead to inefficient use of resources and slower response times to critical customer needs.
• Negative or Sensitive Feedback Delivered Directly: Delivering raw, unfiltered feedback could expose teams to highly negative or even sensitive customer comments that may not be suitable for all employees to process. Without any curation, emotional or harsh feedback might demoralize teams or cause confusion. If feedback includes personal attacks or emotionally charged language, it could lead to team morale issues, stress, or even internal conflict as employees try to navigate harsh critiques without context or moderation.
• Fragmented Understanding Across Teams: While direct feedback may reduce distortion, it could result in a fragmented understanding across teams. Different teams may interpret raw feedback in various ways, leading to inconsistent responses and a lack of alignment across departments. This inconsistency could harm the customer experience, as teams may address issues differently without a unified approach or shared interpretation of the feedback.
• Absence of Data Interpretation: Unfiltered feedback often lacks data interpretation that highlights underlying trends or patterns. Without this layer of analysis, teams might struggle to see the bigger picture or connect related insights. Important trends or systemic issues may go unnoticed, resulting in short-term fixes instead of addressing root causes that affect broader customer satisfaction or product quality.
• Emotional Reaction to Feedback: When raw feedback is delivered without filtering, there is a risk that teams might respond too emotionally, either overreacting to criticism or feeling demotivated by negative comments. Emotional reactions to feedback can result in hasty decisions or a lack of focus on constructive action. Teams might focus more on appeasing vocal customers or overcorrecting based on emotional responses rather than data-driven improvements.

Choosing the Right AI Tool

When selecting an AI tool for qualitative research and insights, consider the following factors:

• Established Track Record: Look for tools that have been tested across industries by both startups and Fortune 500 companies, demonstrating reliability and effectiveness. If you're interested in trying a newer, up and coming tool, see what other Product Strategy, Customer Discovery, UX Research, and Design leaders use and recommend themselves. Some have even started their own companies. I know that I place more trust and credibility in those experts that are solving for problems that they have deep expertise and personal experience with!
• Scalability: Consider tools designed to accommodate the needs of both small businesses and large enterprises. Some niche-focused solutions may lack the robustness required for larger-scale operations.
• Feature Sets: Opt for platforms that offer the features you need, such as research repositories, in-depth analytics, CRM integrations, and/or customizable options.
• Integration & Customization: Perhaps you may need to ensure that the tool integrates seamlessly with existing systems, making it easier to incorporate into larger workflows.
• Security and Privacy: Prioritize tools that comply with industry standards for data protection and privacy, so that customer information and insights are handled securely and ethically.

AI tools have immense potential, however, when evaluating your use of them carefully consider whether these tools can solve deeper, more nuanced customer insight challenges, particularly in sectors where human insight still plays a critical role in decision-making.

Some AI Tools to Consider

Here’s a brief overview of some of the AI tools that I use or have used:

• Hey Marvin: An innovative tool that analyzes customer conversations and transforms them into actionable insights in various formats. It automatically categorizes feedback by topics, making it easier for teams to focus on the most critical areas. **While Hey Marvin is a relatively newer tool that focuses on analyzing customer conversations to provide actionable insights, it's gaining traction among teams looking for efficient ways to process qualitative data, and user feedback suggests it’s effective in transforming conversations into insights. I've been using Hey Marvin for the last few months, and really enjoying it. Thank you Kyle Soucy for the recommendation!
• Dovetail: This tool automates transcription and helps organize and analyze qualitative data, supporting AI-powered tagging and thematic analysis.
• dscout: A tool that allows teams to capture in-the-moment user insights through video diaries and mobile ethnography. It helps researchers collect rich, contextual qualitative feedback, especially useful for gathering real-world experiences and behaviors from users.
• Optimal Workshop suite of tools including Qualitative insights: Uncover valuable perspectives and insights to inform and elevate product development, design decisions, and content and marketing strategies.
• Qualtrics: A comprehensive experience management platform that automates the collection and analysis of customer feedback across multiple channels, delivering AI-driven summaries and detailed insights crucial for large-scale management.
• UserTesting / EnjoyHQ: Offers automated analysis of user interactions and feedback from usability tests, providing deep insights into user behaviour and experiences.

What about having conversations about insights, and asking questions?

A few weeks ago, I noticed a connection of mine enthusiastically post about how blown away they were when they have a conversation with an AI, called Boardy Boardman - an autonomous social AI tool designed to help individuals and organizations connect with relevant people in their network. Like many people, I was a little sceptical but then when I did talk over the phone with this service, it really struck me how good of an application it could be, to not only use it to refine content and practice for presentations, but also to extend the reach and interactivity of research insights and reports. While Boardy doesn’t necessarily focus on consuming reports or facilitating conversations about insights from qualitative research, tools like ChatGPT, and Descript, are currently better suited for tasks like reading reports, tailoring them for different audiences, and engaging in conversations about the content.

There are examples of several AI tools currently available that can consume content, read it out, and then also engage in a conversation about the material. These tools often integrate natural language processing (NLP) and voice synthesis to provide interactive experiences, allowing users to ask questions or discuss the content in real time.

Here are some options:

ChatGPT with Voice Mode (e.g., OpenAI's API with Speech-to-Text and Text-to-Speech Integration)

• You can use ChatGPT (or similar language models) in combination with voice technology to consume text and have conversational interactions about it. By integrating speech-to-text (STT) and text-to-speech (TTS) systems (like Google Speech API or Azure Speech Services), the AI can read content aloud and engage in a conversation about it.

Descript

• Primarily known for video and audio editing, Descript has voice synthesis and AI transcription features. It can read content aloud in a natural-sounding voice, and you can interact with it by making changes or generating discussions on specific parts of the content.
• How to use it: Import content or transcriptions into Descriptm and use its voice generation feature to have the text read out. Then, combine it with other AI conversation engines (such as ChatGPT) to enable interactive Q&A around the content.

Surely it is only a matter of time before the tools that I listed previously allow users to have customized conversations about insights, and ask questions to get tailored answers about reports!

--

In today’s competitive landscape, where speed and scalability are often prioritized, AI tools present a powerful opportunity to streamline data collection, analysis, and reporting. But understanding the deeper 'why' behind customer behaviours still requires the human touch—empathy, context, and nuance that AI simply cannot replicate. Businesses that balance AI-driven efficiency with human-driven insights are best positioned to thrive.

The key is to strike the right balance—using AI to handle the heavy lifting and surface patterns, while relying on human insight to interpret nuance, context, and emotion that machines currently can’t replicate.

As AI continues to evolve, we will likely see even more advanced capabilities that allow us to interact with qualitative data in richer, more dynamic ways—helping businesses refine customer experiences and drive innovation. Businesses and organizations that optimize the best of both worlds—AI efficiency and human interpretation—will be poised for long-term success. By leveraging AI tools alongside qualitative research expertise, we can enhance our understanding to create exceptional products and experiences.

Have you experimented with integrating AI into your processes? Let’s discuss the tools and approaches shaping the future of customer understanding.

💡 What tools have you used that you would recommend, and why?

#QualitativeResearch #AIinBusiness #CustomerInsights #ProductDiscovery #UserExperience #ProductDevelopment #MarketResearch