Franck — The AI Consultant Agent

Franck is an AI-powered conversational agent that conducts structured interviews on your behalf, at scale. By email or by web chat or by audio/avatars. Supervised or autonomous.

Where a human consultant or research analyst can run two to four in-depth interviews per day, Franck supervises fifteen to hundreds in parallel. He asks the questions, follows up, rephrases, detects tone, and delivers a synthesis you can act on.

Franck is not designed to replace human judgment on strategic conversations. He is designed to handle the structured, repetitive part of interviewing - so the people in your organization can focus on what only they can do: synthesis, judgment, and decision.

Some Details !

Why it matters

Scale changes what's possible. Most organizations limit themselves to interviewing ten or fifteen people on a topic — not because that's enough, but because that's all the bandwidth they have. Franck removes that ceiling. Interviewing one hundred or five hundred people on a structured topic stops being a special project. It becomes a standard practice.

Consistency replaces variability. Every respondent is interviewed against the same framework, with the same rigor in follow-up questions. Interviewer bias disappears. Findings are genuinely comparable, across teams, across geographies, across time. Senior time is freed up. The repetitive, time-consuming part of fact-finding — scheduling, conducting, transcribing, structuring — moves out of the consultant's, analyst's, or executive's calendar. They focus on the cross-cutting view, the judgment call, the recommendation.

Cost drops by an order of magnitude. A 45-minute interview run by Franck costs few euros by email or web chat, against forty to one hundred euros for a consultant-led equivalent (preparation, conduct, and write-up included). The economics make consultations possible that simply couldn't be funded before. Languages and geographies are no longer a barrier. Franck operates natively in French and English, and the same project can run interviews in parallel across markets, with comparable outputs.

How it works

Franck runs as a dedicated application, hosted on our server inside NEW LINK PARTNERS. we propose to conduct the interviews on your behalf (fees per interview) but you can have own server located on your premises on your environnement (licence fees). In this case, all data — interviews, contacts, guidelines — stays on your premises. Franck connects to your own email server and to Anthropic's Claude API through your own account, with full transparency on usage and costs.

Three communication channels are available. Email, with each interview tracked through a unique reference, follow-ups handled automatically, and PDF attachments supported. Web chat, accessed by the respondent through a secure personalized link, with messages flowing in real time, or using an avatar and text-to-speech and speech-to-text capabilities.

Two operating modes. In supervised mode, every AI-generated response is presented as a draft for human review before being sent. In autonomous mode, Franck conducts the conversation end-to-end, with the option to escalate to a human if tone analysis detects negative sentiment.

A project-level chat lets you query the entire body of interviews conducted within a project, asking transversal questions and surfacing patterns across respondents.

Use cases (illustrations)

Voice of customer at scale. Interview hundreds of customers in depth on their actual journey — onboarding, support, contract renewal, churn — well beyond NPS scores. Surface verbatim insights across statistically meaningful volumes.
Voice of dealer / partner / supplier. Consult your distribution network or supplier base on relationship quality, operational frictions, and expectations — across the full network rather than a quarterly sample.
Lost prospects and failed conversions. Interview prospects who requested a quote and didn't sign, without commercial pressure, to understand what really blocked the decision. Rare and high-value insight.
Post-incident or post-event listening. After a service incident, a process change, a product launch, or a corporate event, systematically interview customers or employees to surface weak signals that dashboards don't show.
Employee consultation on structuring decisions. Before a reorganization, a system migration, a strategic pivot, gather the structured input of hundreds of employees in weeks rather than months — beyond the annual engagement survey.
Climate and culture studies. Run qualitative engagement studies with real open-ended questions, in multiple languages, with comparable outputs by team, geography, or business unit.
Exit interviews — done properly. Interview every leaver against a consistent framework, surface trends across the organization, identify the real drivers behind regretted attrition.
Reference checks at scale (executive search, M&A). Collect ten or twenty references per candidate or per target, with homogeneous, structured outputs, comparable across files.
Commercial due diligence. In an investment context, interview thirty customers, twenty former employees, and fifteen competitors of a target in two weeks — rather than the eight interviews that typically fit in the actual schedule.
Market and pricing studies. Interview hundreds of prospects or customers on their real intentions (electric vehicles, leasing vs ownership, subscription, alternative channels), with verbatim depth that surveys can't deliver.
Internal audit and compliance interviews. Conduct structured interviews across multiple sites or business units, with full traceability of exchanges and consistent methodology across auditors.
Qualitative market research at scale. For research firms and insights teams, run large-volume qualitative studies — fifty to two hundred in-depth interviews per project — with the comparability and synthesis quality of a hand-coded study, at a fraction of the time and cost.
Cross-portfolio listening. For groups operating across brands, sites, or geographies, run the same study in parallel across the portfolio, with directly comparable outputs.
Stakeholder interviews for transformation projects. For CXO, run the diagnostic interviews that anchor any transformation project — across managers, operations, customers, suppliers — at a scale that matches the ambition of the engagement.

What Franck does not do

Strategic conversations with senior decision-makers. Negotiations. Reading non-verbal cues in sensitive meetings. The interview where the relationship matters more than the data.

That's deliberate. Franck is built to do at scale what today can only be done at small scale, for lack of time. The judgment, the nuance, the human relationship — that stays where it belongs.