The Lewsearch Report

lewn is the free quick-poll chat included with every Lewsearch account.

Ask a thousand people. In ten seconds.

Instead of one AI’s best guess, lewn polls a stand-in panel of hundreds of respondents, modeled on real US Census data across age, region, income & politics, then shows you what they actually think and where they disagree.

Free to try. No account needed to start. Sign up at lewsearch.com →

50 → 250 respondents per questioncompared against real studies

Example · poll result

“How appealing is a 4-day work week?”

Asked 250 US adults · multiple choice

Very appealing58%
Somewhat appealing27%
Not appealing15%
lewn·aiasked 1,000 respondents · compared against real studies

What ChatGPT can’t tell you

ASK CHATGPT

  • —One blended answer
  • —The model’s “average” voice
  • —Disagreement is hidden
  • —Not calibrated to anyone
  • —Subgroups are guessed

ASK LEWN

  • A distribution across hundreds of people
  • Distinct, demographically-grounded respondents
  • Majority and dissent, both shown
  • Checked against real studies (~7.5 pts average error)
  • Cross-tabs by age, region, politics

ChatGPT recalls what’s been measured. lewn predicts what hasn’t.

A new product, price, or message has nothing to look up. lewn simulates a real population so you can read the reaction before you spend. See the proof.

Ready for the full picture?

lewn is the free quick-poll chat included with every Lewsearch account. When you need client-ready research (PDF reports, focus groups you follow across every question, and calibrated panels), the same free account unlocks the full Lewsearch platform.

Sign up at lewsearch.com →

Frequently asked questions

What is lewn?
lewn (lewn.ai) is the free quick-poll chat included with every Lewsearch account. It polls a panel of hundreds of demographically-grounded synthetic respondents and returns the full distribution of opinion: who agrees, who doesn't, and how it splits by age, region, and politics. The aggregate is benchmarked against real polls from Pew, Gallup, the UT/Texas Politics Project, and PPIC.
How is lewn different from ChatGPT?
Ask ChatGPT what an audience thinks and you get one blended answer, its single best guess, with no notion of how many people or which people. lewn samples a population: hundreds of distinct respondents returned as a distribution, with subgroup splits and individual quotes. The aggregate is benchmarked to real survey data (about 7.5% average error on 404 ex-electoral benchmark questions (9.97% on a fully held-out set)). ChatGPT gives you the average voice; lewn measures the room.
How accurate is lewn?
lewn runs on Lewis, the same calibrated model behind Lewsearch: about 7.5% average error on 404 ex-electoral benchmark questions (9.97% on a fully held-out set), versus Pew, Gallup, UT/Texas Politics Project, PPIC, and official election results. Every result shows its own confidence; small-sample or un-benchmarked reads are labeled directional rather than dressed up as precise.
Is lewn free?
Yes. The free tier polls a panel of 50 respondents per question with no signup. Paid Lewsearch plans poll larger chat panels (up to 250 respondents per question) for tighter margins and stable cross-tabs; full studies go up to 10,000.
What can I ask lewn?
How an audience would react to a message, product idea, ad, or policy; which option a demographic prefers; or how opinion splits across age, region, income, and politics. Answered in about ten seconds.
When should I use Lewsearch instead?
lewn is built for a fast, directional gut-check in chat. Step up to Lewsearch (the enterprise platform from the same team) when you need a client-ready PDF report, multi-question studies, panels up to 10,000 respondents, a private custom panel, an embeddable research API with an SLA, or seats for a team. Same calibrated engine: lewn is the quick read, Lewsearch is the deliverable.