---
title: How long does market research take? Minutes per run, under 48 hours end to end
description: A synthetic run takes minutes, and a full study under 48 hours including writing the questionnaire and reading the result. What replaces weeks of field time, step by step.
---

# How long does market research take

A few minutes to interview the profiles, and under 48 hours end to end, writing the questionnaire and reading the result included. That is the timeline of a synthetic study. A fielded study is counted differently: recruiting the sample, chasing it and closing the field take weeks, and that delay is not yours to control.

## Under 48 hours, and what takes the time

| Step | Time | What it depends on |
| --- | --- | --- |
| Write and review the questionnaire | minutes to a few hours | the only step that really depends on you: question order, the opposite version, the closed list of refusal options |
| Interview 100 personas on 10 questions | about 2 minutes 30 | nothing on your side; the population is already built |
| Read the results | from 5 minutes | the time you spend on the gaps, not the availability of the number |
| From a vague idea to settled questions | under 48 hours | a badly worded question shows up here, and is rewritten without waiting for a new field |

That is why an under-48-hour study is still a study, and not a rushed poll: the saved time is not taken from the method, it is taken from waiting.

## What disappears from the timeline

In a classic study, the calendar is set by the field, not by you. Recruiting a sample that matches the target, paying incentives, replacing dropouts and waiting for quotas to fill are the steps that stretch a two-week project into six. None of them disappears because the question is unimportant; they disappear because no one has to be recruited.

### Recruitment

A narrow target, such as buyers of a specific tool in a specific state, is the slowest sample to fill. The panel is already segmented on those attributes.

### Field closure

A field closes when the last quota fills, which is a date you learn, not a date you set. A run finishes when the queue finishes.

### Rewrites

A question that fails is usually discovered after the field. Here it is discovered in the same hour, and the rewrite costs another run, not another two weeks.

## What still takes time, and should

A fast run does not make a decision fast. Three things still take the time they take, and skipping them is how a study produces a number nobody uses.

- **Deciding what would change your mind.** If no result could move the decision, the question is not a research question yet.

- **Wording the question and its opposite.** A scale that only measures agreement cannot tell a strong opinion from a badly framed item.

- **Reading the gaps.** The gap between two versions, two segments or two prices is the finding. The top-line percentage rarely is.

On a small sample, no figure reads as a fine measurement. What stays readable is the sharp gap, which is exactly what a 25-persona free run is for. The [sample size calculator](/us/sample-size-calculator) gives the margin that goes with each sample size.

## When the long timeline is the right answer

Some questions need a fielded panel, and no amount of speed changes that: a share you will report, an incidence rate, a regulated claim, an observed purchase. Those are measured on a sample drawn from the population you address, over weeks, and the cost follows. Synthetic research is the step before, the one that keeps you from paying for a field on a question that was never going to survive contact with respondents.

[Run a synthetic survey](/us/login)

Related: [what a study costs](/us/cost-of-market-research), [research without a budget](/us/market-research-without-a-budget), and [what the method does not do](/us/fiabilite).