Frequently Asked Questions
Everything you need to know about ReadingMinds, from features and technical details to data privacy and getting started.
General & Features
ReadingMinds is an AI-native market research platform designed to save teams hundreds of hours. It is an end-to-end system where researchers can conduct numerous AI-moderated voice interviews to generate insights with the speed and scale of surveys, delivering in-depth analysis within hours. It functions as a research copilot.
ReadingMinds accommodates a wide variety of research scenarios, such as concept, impression, and prototype testing; marketing and PR studies; brand sentiment tracking; and usability and customer experience research. Use cases include consumer trends, new products, messaging & ads, campaign plans, tracking, and employee engagement surveys.
The AI-moderated interviews provide deeper insights than static open-ended survey questions because they facilitate adaptive, emotionally intelligent two-way conversations.
While AI interviews are not a complete substitute for human moderators, they provide significant speed and scale, allowing for more participants and ensuring consistency in probing questions and data quality. Participant feedback is overwhelmingly positive, with 93-97% satisfaction across diverse demographics.
Interviews are typically quick, lasting between 5-15 minutes. The average duration is about 10 minutes.
ReadingMinds can provide comprehensive insights within hours due to its ability to conduct thousands of interviews in parallel with automated analysis.
At launch, ReadingMinds supports American and British English across different speaking styles. Support for more languages is planned for the near future.
Yes, there is a customizable Insights Dashboard. Soon, it will also offer real-time trend monitoring and the ability to set alerts for specific keywords or product mentions.
The platform uses real-time thematic coding, granular sentiment analysis, and entity recognition to synthesize insights from respondent data.
ReadingMinds provides an interactive dashboard, data exports in CSV format, and automatically generated reports that include insights and recommendations.
ReadingMinds uses ASR (Automatic Speech Recognition) for advanced 2-way streaming speech-to-speech communication. It analyzes verbal feedback, performs real-time sentiment analysis, and recognizes prosody to infer sentiment without asking direct questions.
For SMB and mid-market teams with limited time or budget, ReadingMinds offers a way to get in-depth feedback. For large organizations, it serves as a complementary early-stage tool for quick market discovery before larger efforts like focus groups.
ReadingMinds combines quantitative and qualitative methods. It collects data at scale like quantitative research but asks open-ended questions for in-depth answers. Compared to traditional interviews, it is more efficient and scalable, with interviews averaging 10 minutes and the ability to interview hundreds at once.
Consistency, scalability, 24/7 efficiency, unbiased probing, faster turnaround times, and significant cost-effectiveness.
Semi-structured interviews led by an emotionally intelligent AI agent that can ask both open and closed questions and dynamically ask probing follow-up questions.
The ReadingMinds Emotional Fingerprint™ is our proprietary framework for capturing what words alone hide. It classifies six core emotions (sad, angry, confrontational, neutral, cheerful, enthusiastic) and scores each on a 1 to 9 intensity scale. Every insight ReadingMinds delivers is grounded in this analysis, giving you the truth layer beneath what customers say.
The Purchase Enthusiasm Potential score measures emotional energy and readiness to buy in a person's voice. A high PEP indicates high intent; a low PEP suggests low intent or negative sentiment.
Data Privacy & Compliance
Data is stored securely in the cloud on AWS East US. We store transcripts and derived sentiment/emotion signals only; no permanent voice recordings are stored. All storage is compliant with relevant data laws.
All data, both at rest and in transit, is secured using TLS 1.2 or higher encryption with strict transport security protocols.
ReadingMinds is actively pursuing SOC 2 Type II and GDPR certification through Vanta. HIPAA safeguards are in place. Data anonymization is default and no personal data is stored or used for training.
The client is the Data Controller, while ReadingMinds acts as the Data Processor. A Data Processing Agreement (DPA) can be signed.
In compliance with GDPR's "right to be forgotten," ReadingMinds facilitates permanent, irreversible data deletion upon request.
No, customer interview data is kept confidential and is not used to train AI models.
If sensitive information or PII is shared, the AI moves to the next question and automatically strips out such data.
Usage & Logistics
Yes, all collected data, including interviews and insights, can be exported.
Yes, the platform was designed for this from day one.
Yes, the AI voice of "Emma" is watermarked.
Yes, you can cancel your 1-year subscription at any time.
Yes, screener questions can be configured in the Interview Guide for each study.
You can get results in one to two hours. Using panel integration for 100 US general population responses, completion can take about 30 minutes.
If recruiting from your own customer base, prepare the email list. Setting up is straightforward: create a Project, then create a Study with an Interview Guide.
Yes, and this is often desirable. Different answers warrant probing in different areas.
Yes, the AI is trained to recognize such behavior and guides participants back. It uses metadata, AI signals, response pattern checks, digital fingerprinting, and bot-detection.
Yes, you can flag an interview to be excluded.
Yes, this feature is built-in and recommended.
Completion rates have been great. Participants prefer the interactive AI-moderated experience over standard surveys.
While possible, it is not the intended use case. The platform aims for 10-minute average interviews, max 20 minutes.
Yes, you can invite people from inside or outside your organization without extra user costs.
Emma adapts in real-time with dynamic probing. You can specify the number of follow-up questions.
No, the AI stays factual to what participants said.
Yes, each Study generates a unique link you can share with any audience.
AI-driven thematic summaries, key takeaways, auto-coded themes, granular sentiment analysis, transcripts, and insight reports.
Yes, you can export raw transcripts as CSV and all conversations with AI-generated categories.
ReadingMinds empowers, not fully replaces, researchers. The AI automates interviewing and initial analysis while humans provide critical thinking and final insight curation.
Limit of 20 questions per session, with max 3 probing questions each. Recommendation is not to exceed 10.
Minimum of 10 recommended for diverse opinions. For statistical significance, at least 50 interviews are recommended.
Yes, you can create an Interview Guide with your own follow-up strategies or create one interactively with Emma.
Still have questions?
Our team is happy to help. Reach out or take a Live Test Drive to experience ReadingMinds yourself.