People Lie in Surveys. Video Interviews Fix That.

01 /The Concept

Pogo makes a bold claim: it's the only way for brands to have actual conversations with their real buyers.

The process starts with precision targeting. Brands define hyper-specific audience criteria – say, people who bought a particular protein bar brand at least twice from a specific retail chain. From there, they build a questionnaire and deploy Pogo's AI agent to conduct video interviews with those exact people, extracting as much depth as possible.

The output is an AI-synthesized report packed with insights – complete with full interview recordings and transcripts.

Major brands have already run studies on Pogo around questions like: "What turns skeptical riders into loyal Waymo robotaxi users?" "What drives someone to choose one pharmacy chain over another?" "What's the first reaction to Pepsi's new drink?" "Do private-label energy drinks have a real shot at stealing share from name brands?"

Beyond interviews, the platform also lets brands track actual customer journeys – the moments that push or pull buyers toward or away from a purchase decision. For example: find people who walked into a store that carried a brand's product and ask if they noticed it. Or find someone who ultimately bought a competitor's version and ask why they didn't pick the original.

All of this is made possible because Pogo collects rich data on its users – demographics, psychographics, account balances, store visits, purchase history, app usage, subscriptions, investments, and more.

The collection engine is Pogo's consumer app, which tracks location, monitors app activity, and lets users upload digital and paper receipts. Despite that surveillance-adjacent model, 3 million Americans use the app – and Pogo claims it captures data from roughly 1 in every 150 shopping trips made in the US.

The incentive is simple: Pogo pays its users. Every logged purchase action and completed brand interview earns cashback and rewards.

Pogo recently closed a new funding round, though the amount wasn't disclosed. Total capital raised since its 2020 launch stands at $32 million across four rounds.

02 /Why It Matters

Pogo puts its finger on the core flaw in consumer research: most studies draw conclusions from bad data.

People lie – sometimes intentionally, often by projecting what they wish were true. That distortion affects everything from audience selection ("have you bought X in the last 30 days?") to behavioral recall ("why did you choose that product?").

Pogo's founding thesis was to attack this problem directly. The roadmap had three stages: start by building a consumer app with reward mechanics compelling enough to get people sharing honest purchase behavior, even at small scale; then scale that app to a broad US audience to accumulate a reliable behavioral dataset; and now – the stage underway – build a brand intelligence platform on top of that honest data.

That honesty is the competitive wedge that separates Pogo from the crowded AI research platform space.

Other platforms have also adopted AI interviewers, insight extraction, and report generation. The real question isn't which AI is smarter – it's whose data is more accurate. Who you actually interview determines whether the insights are useful.

This keeps reinforcing a theme that comes up repeatedly in startup analysis: for any AI platform, the quality of the underlying data matters more than the sophistication of the models running on top of it.

03 /Opportunities

The most obvious extension: build a Pogo equivalent for a different market. Any geography where brands want consumer insights grounded in real behavior – rather than self-reported surveys – is a candidate.

From there, the question gets more interesting: can niche versions of Pogo work? A platform collecting verified behavioral data within a specific product category or consumer segment seems entirely plausible.

As a reference point, Cafeteria ([related review](/review/kto-pervym-vstal-togo-i-tapki)) raised $6 million to run brand research for companies targeting younger audiences – starting specifically with café and quick-service restaurant shoppers.

Daash ([related review](/review/hochesh-znat-skolko-i-chego-prodajut-tvoi-konkurenty)), with $8.3 million raised, helps beauty brands analyze competitive dynamics through algorithms sharp enough to estimate sales volumes for individual SKUs and explain why specific products gain traction.

Combine verified behavioral data like Pogo's with niche focus like Cafeteria or Daash and the value proposition becomes even stronger.

More broadly, this points toward a pattern: AI platform competition will increasingly shift toward data quality as the key battleground. Which raises the question – in which sectors beyond consumer research can better data already be converted into a competitive moat for a new AI platform?