Frequently asked questions

Candor is a synthetic research platform that generates research-backed personas you can interview before building anything. It combines your uploaded research, web data, and validated behavioral models to create personas with realistic personality traits, cognitive biases, and decision-making patterns. The goal: give product teams honest customer signal, fast, so they know what's worth building before committing resources.

Most AI persona tools generate fictional characters from a prompt. Candor starts with evidence (your documents, web research, validated behavioral data) and builds personas from the ground up. Every trait carries a provenance tag showing exactly where it came from: grounded in your research, inferred from behavioral patterns, calibrated from peer-reviewed distributions, or flagged as low confidence. A critic agent validates every interview response for consistency before you see it.

Candor is built for anyone who needs to validate what they're building. Innovation teams use it when a stage gate is approaching and traditional research would take too long. Consultants and agencies use it to add fast, credible customer validation to their methodology. Startups use it to test assumptions before committing engineering resources. You don't need to be a trained researcher to run a study.

Candor uses a multi-stage pipeline. First, it analyzes your audience description and any uploaded documents to identify behavioral signals. It retrieves corroborating evidence from the web, then generates audience segments based on behavioral and psychographic dimensions, not just demographics. From these segments, it creates archetypes, samples individual personas with OCEAN personality traits and cognitive biases drawn from real population distributions, and generates rich narrative memory for each persona.

Evidence-grounded means every persona attribute traces back to a source. When Candor says a persona has high status quo bias, you can see whether that came from your uploaded customer interviews, from web research about the target market, from peer-reviewed personality distributions, or from inference. Nothing is made up without acknowledgment. Attributes flagged as low confidence are explicitly marked so you know what to trust and what to probe.

Provenance tagging is Candor's system for tracking the origin of every persona trait. Each attribute is tagged as one of: grounded (directly from your research documents), inferred (derived from behavioral patterns in the evidence), calibrated (sampled from validated research distributions like OCEAN personality norms), or weak confidence (assumed but not well-evidenced). This gives you a complete audit trail from any finding back to its source.

Candor offers two interview modes. In live mode, you type questions and the persona responds in real time. It feels like a natural conversation. In auto-interview mode, you configure an interview guide with sections and learning goals, and Candor conducts the interview automatically, tracking coverage and adjusting questions as it goes. Both modes use the same persona inhabitation system, and a critic agent validates every response for consistency.

Candor runs a seven-step synthesis pipeline. It extracts tagged signals from every transcript, evaluates your assumptions against the evidence, clusters findings into themes ranked by frequency and intensity, breaks down how different archetypes experienced the study differently, surfaces genuine tensions where personas disagree, frames product opportunities, and links everything back to source quotes. The result is a structured report designed to drive decisions.

Personas are built on Big Five (OCEAN) personality traits sampled from peer-reviewed population distributions, calibrated by region and occupation. Cognitive biases are assigned with research-backed intensity values. Each persona has narrative memory covering their identity, behaviors, beliefs, and communication style. They remember everything across interview sessions, maintain consistent views, and push back when they disagree. They don't drift into generic AI agreeableness.

Every persona response passes through a critic validation step before you see it. The critic checks for hard contradictions (the persona said something that directly conflicts with an established belief or prior statement) and soft tensions (a notable shift in enthusiasm without a clear reason). Hard contradictions trigger regeneration. Soft tensions are flagged in metadata but still delivered, because real people have variable moods too.

Candor uses the Big Five (OCEAN) model for personality: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. Trait values are sampled from cross-cultural population distributions published in peer-reviewed research. Cognitive biases are modeled as first-class traits with intensity values from 0 to 1, drawing from established behavioral science. B2B and B2C personas use different attribute models reflecting distinct decision-making frameworks.

Yes. Every persona attribute carries a provenance tag. In the persona detail view, you can see which traits are grounded in your uploaded research, which were inferred from behavioral patterns, which were calibrated from validated distributions, and which carry low confidence. The synthesis report traces every finding back through themes to specific interview quotes to the personas and archetypes that generated them.

Candor supports four study types, each with tailored synthesis. Problem discovery explores unknown problems in a market. Problem validation tests whether your hypothesized problems are real and evaluates specific assumptions. Concept testing evaluates a product concept: what resonates, what needs rethinking, and who it's for. Price testing examines value perception gaps, price anchoring effects, and willingness-to-pay ranges.

Yes. You can upload PDF, DOCX, CSV, or TXT files: customer interviews, survey data, market reports, or any research you have. Candor parses, chunks, and indexes these documents, then uses them as the primary evidence source for audience generation. Your documents are prioritized over web research, and signals extracted from them are tagged with full provenance so you can always trace a persona trait back to your original data.

Audience generation and persona creation typically take minutes, not weeks. The exact time depends on the complexity of your audience definition and how many personas you generate. Live interviews happen in real time. Auto-interviews run in the background and complete depending on the number of personas and interview depth. Synthesis runs automatically after interviews and typically completes within minutes.

Yes, and it treats them as fundamentally different. A procurement lead evaluating enterprise software uses a different decision framework than a consumer making an impulse purchase. Candor models B2B and B2C personas with distinct attribute schemas, personality weightings, bias profiles, and buying triggers. You choose your audience type when creating a study, and the entire pipeline adapts accordingly.

Your data is stored in Supabase (hosted on AWS in the US) and is never shared with other customers. Uploaded documents are used only for your studies. Data processing happens through secure APIs with industry-standard encryption in transit and at rest. Candor's full list of third-party data processors is published on the subprocessors page, updated whenever vendors change.

No. Your research data, uploaded documents, and interview transcripts are not used to train any AI models. Candor uses commercial AI APIs for inference only. Your data is processed and returned, not retained for model training. This applies to all third-party AI providers in the stack.

After every persona response, a separate AI model reviews it against the persona's established beliefs, prior statements, seed decisions, and stance change history. It checks for hard contradictions (directly conflicting with something the persona said or believes) and soft tensions (a shift in tone or enthusiasm without a clear reason). Hard contradictions are rejected and regenerated. Soft tensions are flagged but delivered, because real people have variable moods.

Candor samples Big Five personality traits from peer-reviewed cross-cultural research, including Schmitt et al. T-score distributions across 56 nations and Anni et al. occupational personality profiles. Regional and occupational distributions are used when available. When occupational data doesn't exist for a specific role (about 57% of occupations), Candor falls back to regional norms. No traits are assigned randomly.

When you upload research documents, Candor parses, chunks, and indexes them. During audience generation, signals extracted from your documents are tagged as 'grounded' with a reference to the source file. This provenance carries through to persona attributes, interview behavior, and synthesis findings. You can trace any persona trait or report insight back to the specific document that informed it.

Synthetic research is not a substitute for talking to real people. It's a complement. Candor's evidence grounding, personality calibration, and consistency enforcement make its output more structured and traceable than prompting a generic AI. But synthetic personas can't replicate genuine emotional responses or surface truly unexpected insights the way a real conversation can. Use Candor to sharpen your hypotheses and identify where to invest real research budget.

Join the waitlist at runcandor.com. Candor is currently in development and will launch with early access for waitlist subscribers. When you get access, you'll create a study by describing your target audience, optionally uploading research documents, and choosing your study type. From there, Candor handles audience generation, persona creation, and interview facilitation.

Candor is in active development and not yet publicly available. Join the waitlist to be notified when it launches. Early access will be offered to waitlist subscribers first. Candor is built by Highline Beta, based in Toronto, Canada.