POLLS is an AI conversational survey platform provided by curioph株式会社. It sits between traditional user interviews and fixed questionnaires: surveys are distributed via URL, AI follows up based on responses by asking “why,” and the collected customer or employee feedback is then analyzed, structured, shared, and turned into tasks. It mainly addresses common pain points in large-scale research: time-consuming data collection, difficult question design, analysis that depends heavily on experience, and feedback that is hard to reuse across departments.
The product emphasizes human/AI hybrid question design. AI can automatically generate questions, branch scenarios based on responses, and summarize, categorize, further analyze, and report on open-text answers. The site mentions 24-hour automated operation, real-time analysis, and converting results into actionable tasks for teams such as sales, product development, HR, and CS. Its value is not just as a one-off questionnaire tool, but as a way to accumulate multiple studies into an organization-wide voice-of-customer database, supporting before/after comparisons, cross-survey search, and onboarding for new employees.
The official website does not publicly disclose plans, pricing, free quotas, or trial rules; it only provides a contact form. The page uses human interviews with 100 people—around 125 hours and starting from ¥375,000—as a cost reference, but this should not be treated as product pricing. In terms of integrations, the site shows Teams, Slack, Salesforce, Jira, Notion, and HubSpot, and introduces the concept of “workflow generation.” However, the related interface is marked as under development, and API documentation or formal availability has not been disclosed.
The main advantage is that POLLS combines the depth of qualitative interviews with the broad coverage of questionnaire samples, making it suitable for business teams without specialized research skills to collect feedback quickly. Its multilingual capabilities also make it useful for cross-region surveys. Limitations include a lack of public information on the underlying model, accuracy, sample reports, and bias-control mechanisms. Claims such as “unbiased” and the ability to deeply analyze feedback from “1,000 people” still need validation in real projects. Some features are still shown as under development, so enterprises should request a demo and sample reports before purchasing.
The privacy policy is relatively complete: it states that response text, device information, anonymized IP addresses, cookies, and similar data may be collected. Response data is provided to the company conducting the survey. Secondary use, such as model improvement, requires prior disclosure and consent from the company, and is handled through anonymization or statistical processing. Data may be stored on servers in Japan or overseas. Access and payment availability from mainland China are unknown. If you need local alternatives, you can compare 问卷星, 金数据, and 腾讯问卷; if you prioritize globalization and research management, Typeform, SurveyMonkey, Qualtrics, and Dovetail are also worth considering.
⚠ This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on polls.jp official site.
polls.jp is an Japan AI Apps provider. TG4G tracks its product information, an overall rating of 7.0/10, and a China-accessibility score of Workable. Click "Visit Official Site" to reach polls.jp directly.