nexy

Use case

Ask a hundred people a question anonymously

The question arrives without a name on it, and so does your answer going back. One ask, never a panel, and the link between the two is not kept — that refusal is exactly why people answer honestly about what they pay, what they earned, or what went wrong.

Your question goes out without your name on it, and each answer comes back without theirs. The link between a question and an answer is not stored at all, so there is nothing to hand over later — and people answer honestly about the things nobody says with their name attached.

What it looks like

41 answers, and no way to learn which 41. A tool that could tell you who answered could tell somebody else who asked.

How to do it

  1. Write the question the way you would actually say it. The uncomfortable version gets better answers than the diplomatic one.
  2. Say roughly who should be asked: your street, your town, parents at the school, people in your trade.
  3. It goes out once, with no name on it. Each person answers once, in a sentence, and joins nothing.
  4. You get the pattern back: the spread, the number of replies, and anything several people raised unprompted.
  5. If you want to go deeper you ask a second question. You never get a panel you can come back to.

Why it works

Anonymity has to be structuralNot hidden, not behind a permission: the link between question and answer is not stored. That is why the answers are honest.
One sentence, no obligationPeople answer because it costs them nothing and leads to nothing. Asking them to join a panel ends the honesty immediately.
Spread, not an average“€4 to €10 with most under €6” is the answer. An average of €6.80 would hide the whole finding.
Unprompted repetition is the signalSeven people volunteering that the cheap option has a two-year wait is the most valuable line in the result.
It works for the things nobody says out loudPay, rent, what a builder charged, why somebody left a job, whether a school is as good as it looks.

Where it stops

The refusals here are the product. Each one is the reason the answers are worth having.

You cannot learn who answeredNot on request, not later, not for a good reason. The link is not kept, so there is nothing to produce.
No panel to come back toYou get an ask, not an audience. A reusable group of answerers would stop being honest within weeks.
It will not average away disagreementYou get the spread and the number of replies, and you weigh it yourself.
Not a research instrumentIt is a crowd of real people answering once. It is not a representative sample and does not claim to be.

Questions people ask

How can I ask a lot of people a question anonymously and actually get honest answers?

Ask it once, without your name on it, in a way that costs each person one sentence and leads to nothing afterwards. The answers come back without names too, and the link between a question and an answer is not stored at all. That last part is why people tell you what they really pay, what they really earned, or what really went wrong — there is nothing that could later be handed over.

Can I find out who gave a particular answer?

No — not on request, not later, and not for a good reason. The connection between an answer and a person is not kept, so there is nothing to produce. This looks like a limitation and is in fact the entire mechanism: a tool that could tell you who answered could tell somebody else who asked, and the moment that is possible the answers stop being honest.

What kind of questions is this good for?

The ones you cannot ask with your name attached. What people round here actually pay for childcare or rent. What a builder charged for the same job. Whether a school is as good as it looks from outside. Why somebody left a company. What a freelancer in your field charges. All things where the public answer and the private answer are different.

Why do I get a spread rather than a single answer?

Because the spread is usually the finding. “Between €4 and €10 an hour, most of them under €6, and eight above €10” tells you something real about a market. An average of €6.80 tells you almost nothing and hides the fact that there are two different things happening. You also get the number of replies, so you can weigh it yourself.

Do the people answering get added to anything?

No. They answer once and that is the end of it — no thread, no follow-up questions, no list they are now on. That is the deal that makes them answer at all, and it is also why you do not get a reusable panel. If you want to go deeper you ask a second, separate question.

Is this a survey or a piece of research?

Neither, and it should not be presented as one. It is a number of real people answering a question once. It is not a representative sample, it has no methodology, and the people who choose to answer are not a random slice of anybody. What it is good at is telling you the shape of something nobody will say publicly — which no properly sampled survey will get out of people either.

Who exactly gets asked?

Roughly the group you name: your street, your town, parents at a particular school, people working in your trade. It is deliberately rough, because a precise definition of the audience starts to identify the answerers. You get told how many answered, which is the part you need in order to judge the result.

What is the most useful thing in the answers?

Usually something nobody was asked about. When seven people independently volunteer that the cheap option has a two-year waiting list, that is the line that changes your decision, and no question you wrote would have produced it. Unprompted repetition gets put first in the result for that reason.

Related use cases

New here? How Nexy works explains the whole thing in a minute, and the index lists every use case.

Stay posted on Nexy

We add a new use case most days and test what actually works. Get the short version by email, or a notification the moment something ships. No account, and one click to stop.

Ask a hundred people a question anonymously and get honest answers