Direct answers on synthetic research

What is synthetic research?

Synthetic research uses living agents instead of recruiting humans. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them. A screen for research teams. Hire humans when the number has to stand up in court.

Is synthetic research a probability sample?

No. Lewsearch interviews census-grounded living agents. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them. Use it to screen. Hire humans when the number has to stand up in court, at a regulator, or in a newspaper.

What is a synthetic panel?

A synthetic panel is a set of simulated respondents built from census demographics, asked the same instrument, and aggregated like a survey.

Synthetic panels with living agents

Living agents keep a persistent profile and a memory of prior answers. Lewsearch tabulates them like a survey. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them.

synthetic data market research

Synthetic data in market research

Synthetic market research means simulating respondents instead of recruiting humans. Use it to screen questions and messages. Do not use it as a legal sample.

What is silicon sampling?

Silicon sampling uses language-model answers as stand-ins for survey respondents. It is a methods term, not a Lewsearch brand, and NIM did not evaluate Lewsearch.

Best AI market research tool

The test is whether the vendor publishes error rates, against which benchmarks, and at what n.

Best synthetic research panel

A synthetic panel is a census-grounded set of simulated respondents you can tabulate. The best one is the one you can score against real polls.

AI focus groups

An AI focus group seats a small tracked room on top of a full calibrated panel. Every study is calibrated against published survey data before it reaches you. Focus groups are part of client engagements.

AI focus group software

Conversational interview tools and tabulated synthetic panels are different products. Lewsearch is panel-shaped.

Synthetic panel vs AI focus group

A synthetic panel returns option shares you can tabulate. An AI focus group returns a small tracked conversation. Lewsearch ships both, with the panel as the statistic.

Message testing with AI panels

AI panels can rank messages and surface objections before media spend. Directional accuracy is the decision-grade signal, not a sampling margin. Quotes are synthetic, not human testimonials.

AI concept testing

A stated-preference screen on a synthetic panel: comprehension, objections, and choice. Not a sales forecast, not conjoint, not price elasticity.

What is a Concierge research study?

A Concierge research study is a managed engagement: Lewsearch writes the instrument, fields it on a calibrated panel, and delivers the report. Every study is calibrated against published survey data before it reaches you. Pricing is scoped to the audience and the instrument on a short call.

Are AI survey panels accurate?

They can be, within a published error. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them. A narrower April test on 5 places, 10,000 respondents, scored 7.47 points on 404 non-electoral questions.

Synthetic respondents accuracy

Accuracy claims in this category use different units: MAE, Spearman correlation, and thematic parity. They are not interchangeable.

AI polling accuracy

On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them. Every study is calibrated against published survey data before it reaches you.

Synthetic research tools that publish their error rate

A published error rate has a number, a denominator, named polls, and a file. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them.

How much does synthetic market research cost?

Synthetic market research is priced from the audience, the volume of studies, calibration to past studies, custom audiences built from a client's own customers, and the deployment model. Pricing is scoped to your audiences and study volume on a short call. Your first qualified U.S. consumer Blind Challenge is free.

Enterprise AI market research pricing

Enterprise pricing is scoped to your audiences, your study volume, and the deployment. Every study is calibrated against published survey data before it reaches you. Enterprise panels are also calibrated to your own past studies. Pilots run 8 weeks and roll into an annual Enterprise engagement.

Lewsearch engagements: team workspace, Enterprise, and Concierge

Lewsearch engagements are a team workspace, Enterprise, and Concierge. Pricing is scoped to your audiences and study volume on a short call. Your first qualified U.S. consumer Blind Challenge is free. Pilots run 8 weeks and roll into an annual Enterprise engagement.

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.

Can research firms white-label synthetic panels?

Yes. White-label delivery for research firms and agencies is offered. Client teams deliver studies under their own brand. Full white label is part of Enterprise engagements. Pricing is scoped to your audiences and study volume on a short call.

npm package for survey research

The supported integration is POST /api/v1/cohort/query. A package named lewsearch appears on npm under the Swarmgram account; it is not a supported SDK.

MCP server for market research

MCP access is part of every Lewsearch Enterprise deployment, so AI assistants and internal agents can field questions to a calibrated panel directly. The REST contract is POST /api/v1/cohort/query.