What does ‘quality’ really mean in qualitative research today?
It’s a deceptively simple question but right now, one that sits at the very heart of our industry.
With AI-generated responses on the rise, synthetic respondents quietly creeping into datasets, and participant fraud making headlines, the foundations of trust in research are being shaken. According to the Global Data Quality (GDQ) initiative, nearly 40% of records might be problematic in some way – with 4–5% directly linked to fraud. It’s a sobering figure.
And it’s not just about fraud. Much of the damage comes from something more subtle: bored, disengaged or confused participants. Poorly written surveys, lengthy forms and unengaging questions lead to inattentiveness – ultimately skewing results and undermining insight.
As MRS Managing Director, Debrah Harding, pointed out in a Research Policy & Standards webinar for June 2025: “We’re not just fighting fraud – we’re fighting fatigue.”
So, where does that leave qualitative research, which relies so heavily on nuance, authenticity, and depth?
Why the definition of quality is changing
Market research is facing what some are calling a “data quality crisis.” And rightly so. A growing body of evidence suggests that participant quality is declining not just because of fraud, but due to inattentive respondents, poor survey design, and the commoditisation of data collection.
At the MRS Annual Conference, Debrah Harding discussed how synthetic data is not the escape route from the data quality crisis.
And yet, automation is exactly what many are leaning into. At the same time, downward pricing pressure means some suppliers are cutting corners on validation to stay competitive, a move that seriously compromises quality.
It’s why organisations like MRS and AQR have joined forces with the GDQ to push for clearer, more consistent standards around research quality. As Harding also pointed out, this is about building consistency, transparency, and confidence, so buyers can make informed decisions and trust the data they receive.
Fraud, AI and fatigue are changing the challenge
AI has added a whole new dimension to the challenge. From automated bots to ChatGPT-generated open-ends, AI can now convincingly mimic human responses making it harder to identify fakes with surface-level checks.
In fact, Quirks recently described the wave of participant fraud in qual as a “wake-up call”. It's not just survey speeders or duplicate accounts anymore it's fabricated personas, synthetic voices, and fully AI-written responses. And often, the tools we’ve relied on for years just aren’t catching them.
All of this is contributing to an environment where trust is harder to earn and easier to lose.
For qualitative research, though, quality is about more than simply keeping fraudulent participants out. It means recruiting people who genuinely meet the criteria, understand what’s involved, are engaged in the process and can contribute meaningfully to the research.
Why data quality is business-critical
Poor data isn’t just a technical issue it’s a commercial risk. A 2024 Research Live feature highlighted how quality lapses can lead to misguided strategies, lost revenue, and reputational damage.
As the article put it:
“Without reliable data, even the most sophisticated analytics will produce flawed insights. Quality is not a nice-to-have; it’s business-critical.”
And in qualitative research, where every participant’s voice shapes the story, the stakes are even higher. One poor respondent can skew the direction of an entire project.
How the GDQ is helping to raise standards
The Global Data Quality initiative isn’t just another framework. It’s an international collaboration aiming to introduce clear, evidence-based standards and tools for buyers and suppliers.
Among its goals:
- Define consistent benchmarks for data quality
- Help buyers ask the right questions and identify warning signs
- Provide a standardised glossary, helping to align language across the industry so buyers and suppliers can speak the same data quality language
The GDQ's benchmarking study, now in its second published wave (H1 2026), lets businesses compare their own performance against industry norms.
For a deeper look at the wider data quality challenge, the latest benchmarks and the industry response, explore our market research data quality guide.
Which brings us to the most important point…
What should you look for in a qualitative fieldwork agency?
In this climate, a reputable fieldwork agency isn’t just helpful – it's an important part of protecting research quality. But not all agencies are created equal.
1. Robust, human participant validation
At Angelfish, every respondent is validated in-house, by phone, by a RAS-accredited member of our team. This isn’t just a courtesy call, it’s a crucial step to make sure people genuinely meet the criteria, understand what’s involved and are able to contribute meaningfully to the research. It also helps filter out synthetic or low-quality respondents before they ever make it into a session.
Not all agencies offer this. Some outsource recruitment or rely solely on digital profiling and form-based checks which leaves you open to error, misrepresentation, or worse.
2. Active community management
We review our opt-in community weekly to remove duplicates, flag suspicious activity, and clear out inactive accounts. Our participant database isn’t a static list it’s a living, breathing community that’s constantly being refreshed and refined.
3. Transparent recruitment and strong ethical standards
Angelfish is proud to be fully RAS-accredited, an MRS Company Partner, and certified under the UK Government’s Cyber Essentials scheme, demonstrating our commitment to secure data handling and protection across all projects. We adhere to the highest ethical standards, including GDPR compliance, transparent consent, and clear participant communication throughout. We’ve also signed the Global Data Quality Excellence Pledge, supporting shared standards and continual improvement across the industry.
4. Human-first, not AI-first
Technology can support the process, but it doesn’t replace human judgement. We embrace tools where they add value, while keeping real conversations, experienced oversight and careful validation at the heart of recruitment.
5. A participant experience that supports engagement
We don’t just recruit, we build relationships. By building relationships with our participants, we reduce dropouts, boost engagement, and create a better experience for both researchers and respondents. As ESOMAR notes, participant experience is directly linked to data quality and it’s something we take seriously.
Looking ahead: quality as a collective responsibility
Data quality isn’t something a single agency or organisation can solve on its own. It’s a shared responsibility and initiatives like the Global Data Quality benchmarking study, which has since published two full benchmarking waves covering the UK and other major markets, will be instrumental in setting the standards we all need to live by.
But in the meantime, we believe fieldwork agencies like Angelfish can play an important role. Through transparency, rigour, and a human-first approach, we’re helping to raise the bar for qualitative research, one brilliant respondent at a time.
As part of this commitment, Angelfish Fieldwork is proud to have signed the GDQ Pledge, demonstrating our support for collaborative standards and a shared focus on improving data quality across the industry.
Ultimately, good qualitative research still comes down to trust: trust in who has been recruited, how they have been validated, and the experience they have throughout the research process.
Interested in working with a fieldwork agency that puts data quality first?
Get in touch – we’d love to show you how we do it.
-3.png)
