What a market-research platform owes its participants

Market research runs on people who answer questions honestly for a modest reward. Everything downstream, from the incidence assumption in a proposal to the chart in a board deck, rests on that. Yet participants are the party in the chain with the least information, the least leverage and, on many platforms, the least respect. This post is about what we think a platform owes them, and how those obligations show up in the way mypinio is built.

A reward stated before the work, and paid as stated

The first obligation is the plainest. A participant should know the length and the reward of a survey before opening it, and the reward should be paid when the completion is confirmed. Bait-and-switch rewards, "up to" amounts and rewards that quietly shrink after the fact are the fastest way to teach participants that the platform cannot be trusted, and a participant who distrusts the platform stops caring about the answers.

At mypinio every survey shows its length and its reward in Pintos before it starts. The reward is booked the moment the survey is completed, held visibly while the buyer confirms, and settled at the amount stated. The exchange rate to money, the cash-out minimum and the referral share are published on the rewards page as configured today, not as marketing round numbers.

No personal data as the price of admission

Platforms fighting fraud reach for identity: phone verification, ID uploads, selfies, bank details. Each of these shifts a cost from the platform to the participant, and each creates a store of sensitive data that exists only because the platform could not tell real people from fake ones any other way.

We took the harder route. Identity at mypinio comes from device and behaviour signals: whether an account behaves like one real person, whether surveys get real attention, whether answers are consistent over time. No phone number is verified, no document is uploaded, and the account holds an e-mail address, a name and profile answers. Survey answers reach buyers anonymously and in aggregate. A participant can export everything we hold or delete the account from the privacy section in the app at any time. This is more work for us than a phone check, and it is the right trade.

A second chance before a consequence

Quality checks make mistakes. A participant who pasted an answer from a notes app, or who remembers what they wrote to a similar question last week, looks like a copy-paste operation to a check that only sees the text. A platform that hard-blocks on a single signal will, over time, remove a great many honest people without ever telling them why.

Our open-text check before a study asks once more, with a hint, when an answer looks empty, copied or generated. An honest answer on the second attempt passes and nothing else happens. Only a second failure ends the session, and only repeated patterns lead to a pause. No account is paused without a person looking at it, and every pause can be appealed in writing and is answered in writing. The principle is simple: a consequence should follow a pattern, not a single moment.

Transparency about what is checked and why

Participants are entitled to know the rules they are judged by. Not the thresholds, which would only help the people gaming them, but the categories: that one person may hold one account, that attention matters more than speed, that open-text answers are read for effort, and that rushed or contradictory completions may be rejected by the buyer and reversed. Our safety and fairness page says exactly this in plain language, and the hub guides explain matching, screen-outs and the holding period in the same terms. A participant who understands why a survey ended after three questions does not feel cheated by it.

Honesty about the limits

The most common breach of trust in this industry is not fraud; it is overpromising. "Earn money from home" copy that implies a wage, sign-up bonuses that cannot be cashed out, brand catalogues that turn out to be empty in the participant's country. We would rather say the uncomfortable things ourselves: that screen-outs happen, that some weeks are quiet, that a reward can be reversed when a buyer rejects a completion, that survey income is extra income and not a job. Participants who join with those expectations stay; participants who join on a promise leave angry, and they are right to.

Support from a person, in writing

A platform that pays people owes them a way to raise a problem and get an answer. At mypinio support is reached from inside the app, every message is answered in writing by a person, and decisions about pauses and reversals are explained rather than announced. Writing, not calls, because a written answer can be read again, forwarded and checked against what the page said.

Why buyers should care

None of this is charity. A participant who is paid as stated, checked without being surveilled, given a second chance when a check misfires and told the truth about the limits is a participant who answers carefully and comes back. That is the participant every buyer says it wants and few platforms are built to keep. Data quality is not produced by catching bad participants after the fact; it is produced by treating good participants well enough that they stay. The fairness policy and the quality layer are the same project seen from two sides.

The honest limit

We do not control the questionnaires buyers field, the reasons they reverse or the number of studies looking for a given country in a given month. What we control is the part in between: the rules a participant is subject to, how those rules are explained, how mistakes are handled and what is asked of them in return. On that part the obligations above are the standard we hold ourselves to, and the standard participants should hold us to.