Frequently asked questions

Zeitgeist Labs is a Berlin-based institute for applied social and health research. We develop new research methods and technologies to understand the lived experiences of more people, in less time and with greater depth. We conduct research for foundations, academic institutions, public sector organisations and businesses, while also carrying out our own studies on important societal issues. By doing so, we bring perspectives to light that often go unheard and help give them a stronger voice in important decisions.

We work to the same standard: rigorous methods, experienced researchers, and we stand behind the result. The difference is in how we collect the data. With AI-powered interviews we can talk to everyone rather than a small sample, in their own language, in days rather than months. Your data also stays useful after the report. You can keep asking it questions and add new sources.

We see a project through from the first question to the report. We do evaluations and impact measurement, research alongside programmes and pilots, health services research, and needs and audience analyses. We also survey participants, employees and stakeholders. Our services page gives an overview.

Our clients have one thing in common: a community whose knowledge and perspective they want to understand. That might be voters in a region, participants in a study or programme, a company’s employees or patients in a health system. Our References page shows some of them.

Yes. We have our roots in research and remain active in the academic community. We regularly co-write proposals, as a project partner or as the team that runs the data collection.

Surveys give us numbers. Interviews help us understand what lies behind them. Through conversations, we learn more about the experiences and perspectives behind people’s answers. Until now, conducting these interviews with large numbers of people was often too time-consuming and costly. AI-powered interviews make this possible at scale. They don’t replace interviews conducted in person, but add a new method to the research toolkit. This allows us to collect new types of data, gain insights that were previously difficult to obtain and combine different research methods where it makes sense.

Yes. In studies, trained reviewers could not distinguish AI interviews from human-led ones, and participants preferred the AI interviewer. Answers in an AI interview are about two and a half times as long as answers typed into an open text field. What matters is that the interviews are set up properly and the results are checked, so our researchers always do both themselves.

No. Our researchers develop the research questions, design the studies, and are responsible for scientific quality and interpreting the results. AI doesn’t replace this work. Instead, it expands what we can do in research. It allows us to have in-depth conversations with large numbers of people, analyse vast amounts of different types of data and uncover new patterns and connections. This means our researchers can not only work faster and more comprehensively, but also ask new questions and gain insights that were previously difficult to obtain.

We don’t simply rely on the output of an AI model. Our researchers develop proprietary methods and technological systems grounded in established scientific standards and best practices from social, behavioural and health research. Through systematic quality checks, we assess the consistency, transparency and validity of our analyses. This allows us to combine the capabilities of modern AI with scientific rigour, providing a reliable foundation for research and informed decision-making.

Yes. We tailor each interviewer to the topic and the target group. In one study our interviewer spoke with front-line doctors about the standard of care in breast cancer. In another it captured, in detail, what a company’s experts know about its complex technical systems.

No. Depending on the question, we combine AI interviews with structured questionnaires and interviews led by our researchers, and we draw on other research, documents and existing data. Our researchers work out the right combination with you.

With a 30-minute call about what you want to find out. We listen, discuss methods and the study design with you, then set up and test the AI interviewers, and data collection starts. You can commission us by direct award, under a framework agreement or through a tender.

Not much. You bring your question and, if you have them, contacts for the people you want to hear from. We do the design, the data collection and the analysis, and agree the important steps with you.

Usually our clients already have access to their target group. The interviews work in more than 20 languages, by phone, and without any reading or writing, so we also reach people whom conventional surveys often miss. Nobody needs an appointment: participants take part when it suits them, at home, on the train or in the car. We are happy to advise on invitations and incentives.

Data collection can start at very short notice, and you will not wait months for a report. As soon as the first interviews are in, you can see the results in the Intelligence Dashboard as they arrive.

That depends on the scope, the complexity and the number of interviews. Our studies usually cost considerably less than comparable research done with human interviewers. There is no fixed price per participant and no subscription. The Intelligence Dashboard is included for the length of the study and for several weeks afterwards.

Yes. Our researchers build your framework, for example your impact model, into the Intelligence Dashboard, so you see the results in the form you already know.

Yes. You can upload reports, presentations, strategy papers or interviews from earlier studies to the Intelligence Dashboard. We also bring in data from other internal and external sources, such as existing research, documents and data sets, and compare it with what the conversations show. Everything can then be analysed in one place.

We comply with the GDPR, and the data is stored on servers in the EU. We do not train models on your data, and we use it only for the research purpose agreed with you. Our privacy policy has the details on hosting, storage and deletion.

You do. The data and the results are yours, and you decide how they are used and published.

No. Answers are anonymised, and the client cannot tell who gave them. The one exception: studies where participants are told in advance that their answers will be attributed to them.

Taking part is voluntary, and you can stop at any time. We anonymise your answers. The client cannot see who said what unless you have been told in advance that your answers will be attributed to you. What we store, and for how long, is explained before every interview and in our privacy policy.

In most studies, yes. If an AI conducts your interview, we tell you clearly before it starts. Where the research question calls for it, our researchers conduct interviews themselves. The conversation is recorded and transcribed, and it is always our researchers who analyse it.

For questions or complaints, email hello@zeitgeistlabs.ai. To ask what data we hold about you, or to have it deleted, email privacy@zeitgeistlabs.ai.