About Besample

Built by researchers who spent years catching bad data after the fact

Bootstrapped in Brooklyn. Focused on the quality gap in emerging-market survey fieldwork where straightlining and geo fraud are endemic and standard tooling goes silent.

The team

Elena Brandt

Elena Brandt

CEO and Co-Founder

Elena has spent her career running quantitative fieldwork across MENA and South Asia, designing surveys and managing field quality in markets where standard panel tooling falls short. She co-founded Besample after rebuilding too many toplines following wave-close fraud discoveries.

Marcus Adeyemi

Marcus Adeyemi

CTO and Co-Founder

Marcus focuses on NLP and machine learning applied to behavioral data, with a background in building data validation pipelines for high-stakes data quality contexts. He designed Besample's signal architecture to be platform-agnostic, so it works regardless of which survey platform a client runs.

Yuki Tanaka

Yuki Tanaka

Head of Research Science

Yuki's work is in survey psychometrics and response-behavior research, studying how answer patterns and timing distributions differ across question types and panel populations. At Besample she owns the signal model and threshold calibration framework, adapting detection rules by panel region and question type.

Why we built this

  • Founded 2025, Brooklyn NY
  • Funding Bootstrapped
  • Focus Emerging-market survey quality
  • Contact [email protected]

The problem is not that research agencies are careless. The problem is that standard quality-control steps happen after fieldwork closes, which means every flagged row discovered in analysis arrived too late to avoid reopening quotas or rewriting the deck.

We started Besample because that post-wave fire drill is fully preventable. The signals that reveal inauthentic responses are present in the raw submission data: timing is implausible, answer patterns are monotone, declared geography does not match IP. None of that requires a human to spot it. It requires a system that looks at every row as it arrives instead of waiting until the dataset is complete.

Our focus is emerging markets because that is where the gap is widest. Panel fraud rates in certain MENA and Southeast Asian country pairs are measurably higher than in the markets most quality-tooling vendors calibrate for, and the major platforms do not offer region-specific thresholds. We do.

Get in touch

68 Jay Street, Suite 501, Brooklyn, NY 11201

Ready to run a pilot?

Tell us about your current fieldwork setup and we will scope a free audit together.

Request Access