The Past, The Present & The Possible of AI in Insights

The turn of the year invites reflection - not just on what has changed, but on what is fundamentally shifting. In consumer insights, that shift is now unmistakable. AI has moved from novelty, to an efficiency tool, to something far more consequential: a catalyst for rethinking how organizations build understanding over time, maintain relevance, and make more confident decisions grounded in real-life context.

What began as a drive for speed and cost efficiency is evolving into a deeper transformation. AI is no longer just changing how research is executed, but also what kind of understanding is possible — and how that understanding compounds and travels through organizations.

The real question is no longer whether AI belongs in insights, but whether insights leaders are ready to use AI to move closer to human context and truth.

This shift can be understood across three horizons: (1) the past, where AI enabled qualitative research at scale; (2) the present, where AI augments human researchers to deliver authentic, contextual understanding at the speed of business; and (3) the possible, where AI-powered platforms evolve into continuous insights eco-systems for organizational learning and decision support.

1. The Past: AI-Moderation and Qualitative Research at Scale

Quantitative research has been automated and scaled over a decade ago, while qualitative research has long been constrained by time, cost and human bandwidth. Early applications of large language models in insights focused on breaking these constraints and solve operational challenges. AI moderation made it possible to run many more conversations, adapt follow-up questions in real time, and generate insight faster than traditional one-to-one interviewing ever allowed.

This first wave of innovation was largely operational in nature. The promise was clear: faster studies, lower costs, broader reach. Many early tools delivered meaningful gains in two of three dimensions - speed, cost, and quality - and opened the door to agile, exploratory, and experimental research designs.

Yet these tools also had clear limits. Many of the first applications focussed on automating open ended questions and infused video into quantitative research. Others relied on AI chat and text-based interactions, offering more nuance than open-ends in surveys but missing the observational and non-verbal richness that makes qualitative research powerful. Some employed end-to-end video interviews but were limited to small sample sizes due to processing and analysis challenges.

These tools were not yet advanced enough to replicate human-led conversations for mass qualitative research. Most importantly, insights were still generated in isolated projects, rather than building toward cumulative understanding.

Of course, this phase mattered. It proved that AI could handle core elements of qualitative work - and it laid the foundation for a more ambitious shift.

2. The Present: Scaling The Human Insight Advantage

Today, the impact of AI has moved beyond automation and efficiency. The promise of GenAI in consumer insights goes far beyond simple “AI moderation” or “qual at scale.” The real transformation lies in delivering scaled human insights at the speed of business, while maintaining richness, context, and a human-centred perspective.

This translates in 2 ways:

(a) AI scales your team without scaling your team

The Conveo platform, for example, acts as an expert AI collaborator across end-to-end research workflows:

• Smart study design: AI co-creates discussion guides in minutes, drawing on decades of research expertise and recommending techniques tailored to objectives rather than simply filling templates.
• Human-like video interviews: AI-led voice and video conversations probe deeply, retain context, and scale across cultures, time zones, and languages, enabling hundreds of interviews efficiently.
• Interactive insight generation and co-creation: Researchers can “talk to their data” to surface themes, build personas, generate marketing assets, produce strategy summaries, and answer business questions in real time.
• Compounding knowledge: Each study contributes to an evolving knowledge base, enabling insights to build iteratively over time. Conveo acts as guided intelligence, scaling human expertise rather than replacing it. This iterative approach is the embryonic form of a continuous, compounding model of insight generation - a clear evolution from the one-off, quick-turnaround studies of the past.

By embracing AI as a collaborative partner, insights teams can reinvest the time it frees up into synthesis, sense-making, and advisory work - the activities that most amplify their impact. AI does not replace the insights professional - it exposes where their real value lies.

(b) AI goes beyond the conversation as video captures real behavior

One of the most differentiating capabilities of Conveo’s AI-video moderation and analysis is the ability to combine conversation and observation at scale. AI-led video interviews don’t just ask better questions — they capture non-verbal tone, hesitation, emotion, actions, environments, and behaviour as it unfolds.

Our unique video content analysis bridges one of the longest-standing gaps in insights: the distance between what people say or state and what they do or actually live. It allows insights executives to ground strategic decisions in human context, not abstract averages.

Consumers are no longer removed from their lives and placed in artificial research settings. They show their homes, their shopping journeys, their routines, their products, their screens. The result is insight that feels tangible, credible, and difficult to ignore in decision rooms.

3. The Possible: AI-Powered Insights Eco-Systems

Looking ahead, there is a clear future trajectory. As AI-powered research platforms mature, they will not just support individual studies - they will become the primary environments in which insights work happens.

These platforms will evolve into ‘central nerve systems’ of organisational learning and customer centricity. Each study will add depth, context, and memory. Past insights will inform and generate new questions. Knowledge will compound rather than reset.

Over time, insights teams will work inside living and self-sustaining eco-systems that, e.g.:

• Links historical learnings to emerging questions
• Adapts and execute methods dynamically
• Supports deeper analysis without leaving the system
• Integrates seamlessly into broader organizational workflows

In this future, insights are no longer episodic. They will be continuous, contextual and holistic, and embedded in decision-making. Over time, AI platforms will become trusted research systems and long-term knowledge infrastructures within organizations.

Closing Thought

AI will not replace insights teams.

AI will expose the difference between those who execute research and those who shape understanding and impact business.

The opportunity for insights executives is not only to run more studies faster — but to deliver something organizations have always needed and rarely had: human truth, grounded in reality, at the speed and scale of modern decision-making.

The future belongs to insights leaders who are ready to step into a more strategic, influential role — and use AI to get there.