The Future of the Insight Community is Smaller, Smarter, and Human
By FlexMR
- article
- Agile Qualitative Research
- Agile Quantitative Research
- Long Term Communities
- Online Focus Groups and Forums
- Qualitative Research
- Short Term Communities
- Survey Panel
- Survey Research
- Video Research
- AI
- Artificial Intelligence
- Insight Communities
Summarise with AI
There is much debate about the impact of artificial intelligence (AI) on market research. Last year, it was reported that over 95% of insight professionals have already incorporated AI tools into their work. It is no longer a question of whether artificial intelligence will disrupt, but how.
A HarrisQuest report finds the most expected shift is towards a division of labour where AI takes the wheel – drafting surveys and reports, managing projects, generating synthetic data and processing results. In this near future, technical execution becomes less central to the research skillset than judgment, context, and storytelling.
But AI is not the only disruption the research sector has to contend with. A well-documented data quality crisis, driven by a potent mix of fraudulent respondents, bots, and inattentive participants, continues to erode trust in market research. And all of this is happening at the exact time that we are reckoning with the debate over whether we are truly that far away from a future in which the artificial outnumbers the human.
In an AI-driven world, there is a risk decision makers don’t fully understand their customer, overlooking the humans behind the data points. As Ellie Osbourne, Senior Insight Manager at Saga, explains, “There’s a danger that AI and the growing use of synthetic audiences mean people are seen as data points, rather than the humans they are, but it cannot replace the richness of the human stories behind the data.”
The risk will likely further compound as decision makers are attracted to the scale of data and speed with which business questions can be answered. But as Dr Mark Thorpe cautioned in an article for the MRS Delphi Group “Data should represent people, not replace them”.
Yet a tension remains. Because as organisations deploy models, agents, and intelligent systems, they will need reliable data to train, test, and challenge them. A lot more data. Reflecting on the experience this led to in a previous role, MRS CEO, Jane Frost, warns, “Chasing after more and more data, without quality or purpose, wasted a lot of time, consumed a lot of energy, and led us down blind alleys that took investment that could be better applied elsewhere”.
The response must be to focus on quality of data over quantity. At the end of the day, decision-makers won’t remember data points. They will, however, remember what insight professionals engineer to be memorable. The stories and experiences that they can empathise with. Ellie highlights the importance of this: “Bringing the data to life through human stories and experiences helps decision-makers build empathy alongside evidence”.
The Case for Insight Communities
Insight communities are a long-term continuous research space that offer the opportunity to reach a deeper understanding of customers, surface human stories and generate evidence with empathy. Members exchange views, experiences and needs, enabling organic, evolving insights. They bring people to life through video, forums, scrapbooks and focus groups. In short, you get a fuller picture of your customers, especially compared to one-off surveys.
As Paul Griffiths of Client Advocates explains, “Communities offer a powerful source of the human signal organisations will need. They provide access to identifiable people, not bots, bad actors or synthetic noise, who can be creatively engaged over time to share honest, unfiltered perspectives with clear and transparent provenance.”
Insight Communities in an AI Future
In the future, communities will be even more central to the research mix of brands. Because they are long-term methods, communities follow the same people, deepening understanding over time. The database effectively becomes a full 360-degree view of what customers want, how and why.
It is this rich and evolving dataset that makes them essential in an AI future. AI needs high-quality human signals. Communities provide exactly that. To give that signal impact, the future of communities will be smaller and more engaged, formed of around 100 – 500 members. These members will be heavily profiled customers that build a layer of human-verified truth that sits beneath the trained AI-scale.
Three Workflows Powered by Real Human Verification
In this vision, high-quality, deeply profiled communities provide the foundation for research augmented with AI without diluting its integrity. In essence, the role of the community is as data infrastructure: a trusted, verifiable source of human activity that AI-supported insight is based on.
1. The Living Persona
The real power of the community is the real human feedback, and being able to connect decision-makers to see real people in real time matters.
Yet decision-makers often need speed and scale. A digital persona – an AI simulation trained on the activity of deeply profiled community members – can take over from the human and help decision-makers interact with their target audience on-demand.
2. Organic Insights
When social media arrived at the turn of the millennium, so too did the promise that brands could listen in on the conversations of consumers, uncovering insight that was free from the influence or bias of researchers.
Three dimensions of next-gen communities come together to resolve this challenge and fulfil the promise of organic insight. First, they are comprised of authentically engaged members. That means customers feel valued and contribute to conversations without being prompted. Second, they are profiled based on brand-relevant segmentations (potentially with IDs that are connected and enhanced with POS, CRM and digital tracking metadata).
Finally, AI is able to monitor activity and produce regular high-level thematic summaries in a fraction of the time that humans can.
3. Full surveys with Digital Twins
A digital twin is a synthetic representation of a real person. Rather than a general abstraction of a human generated through volume, a twin is born from the data of a single individual. As Annette Smith, Product Owner at FlexMR, explains, “To create a digital twin you need to do lots of profiling work and capture a broad range of attitudes. Only then can you create an AI simulation capable of answering surveys on their behalf.”
By training AI on highly engaged community members, it’s possible to always return fully answered, perfectly profiled survey datasets without fatiguing customers. To a sector caught in a data quality crisis, the benefit is clear.
This is exactly what Sage has been experimenting with. Eddie o Brein explains, “We ran a test earlier this year. The closest version of the truth wasn’t the agency research panels study. It was the community and synthetic combination.”
In Conclusion
Given the increasing risks of data quality undermining the value of research and the ease with which decision-makers can access high volumes of that data, the role of a community as a quality input will only grow. AI is both a threat and opportunity. It could replace research entirely. Yet it equally creates opportunity, underpinned by human-verified communities to bring data to life and to train AI at the same time.
As Paul Griffiths concludes, “If AI is going to help organisations make better decisions at greater volume and speed, it will need trusted human signal at its core. For the insight industry, that is not a threat to our relevance. It is an opportunity to redefine it.”
