Answers

How long does a synthetic study take?

A queued Lewsearch study returns in about 15 minutes. Every study is calibrated against published survey data before it reaches you. Focus groups, digital-twin interviews, and multi-market studies run on the same calibrated panel. Client teams use a workspace built for their audiences and question library.

About 15 minutes, queued

You submit the instrument and a market with the team. The job sits in a queue. The published turnaround is about 15 minutes. That clock includes a cold start when the model has been idle, the draw of respondents, the interviews, and the report. If the queue is busy or the model is starting, the run can take longer.

What comes back is a tabulated panel: option shares, crosstabs, verbatim quotes, two analyst memos, and a client-ready PDF that says the answers are AI-generated. Coverage and live panels are on coverage.

What those minutes are spent on

You and the team write the questions and pick a market. The system draws simulated respondents whose demographics match that market's margins. Each row is a distinct age, education, income, and party draw. Every study is calibrated against published survey data before it reaches you. The calibrated 50-state benchmark: 7.07-point average miss on the best 80% of 1,521 scored questions (1,216 retained). 10.02 points across every scored question. A narrower April test on 5 places with 10,000 respondents scored 7.47 points on 404 non-electoral questions. The 50-state file uses 200 simulated respondents per question. The artifact is the panel. The statistic is the option share.

What changes the clock

A single-market panel study is the about-15-minute queue. A larger draw, up to 10,000 simulated respondents, is for stabler crosstabs and stays on that queue. A focus group seats a room on the same panel. A digital-twin interview uses the same path. A multi-market study, or a Concierge report the team writes, is scoped with you and takes longer than one queued market. Pricing is scoped to your audiences and study volume on a short call. Book a demo.

Turnaround by deliverable
DeliverableTurnaround
Queued panel studyAbout 15 minutes once queued
Larger panel, up to 10,000 respondentsSame queue, for stabler crosstabs
Focus group on the panelSame queue. The room is seated on the panel.
Digital twin interviewSame queue
Multi-market studyScoped with the team. Longer than one market.
Concierge reportWithin 48 hours of an approved scope

A workspace for the client team

Client teams get a workspace built for them: their audiences, their question library, and their brand. Message tests, including live A/B reads, run from that workspace. REST API access is set up with the team. MCP access is part of every Enterprise deployment and is configured per workspace. The queued study on the calibrated panel is the run that returns in about 15 minutes. Book a demo.

FAQ

How long does a synthetic study take?
A queued Lewsearch study returns in about 15 minutes. Every study is calibrated against published survey data before it reaches you. Focus groups, digital-twin interviews, and multi-market studies run on the same calibrated panel. Client teams use a workspace built for their audiences and question library.
What happens during those 15 minutes?
The study is queued. The model starts if it has been idle. Respondents are drawn to match the market you picked. Every study is calibrated against published survey data before it reaches you. The calibrated 50-state benchmark: 7.07-point average miss on the best 80% of 1,521 scored questions (1,216 retained). 10.02 points across every scored question. The report comes back as option shares, crosstabs, quotes, and a PDF.
Does a larger panel take longer?
A larger panel is for stabler crosstabs. The run stays on the same queue. Focus groups and digital-twin interviews use that queue. Multi-market studies are scoped with the team and take longer than a single-market run. Pricing is scoped to your audiences and study volume on a short call.
Can I get a result the same day?
Yes. A queued study returns in about 15 minutes, including a cold start when the model has been idle. If the queue is busy, it can take longer. Client teams run the next study from a workspace that already holds their audiences and question library.

Check the work

Related answers

Canonical: https://lewsearch.com/answers/how-long-does-a-synthetic-study-take