AI tool for researchers, designers and product teams
Research Synthesis
From interviews to decisions. Upload the transcripts from your qualitative interviews, concept tests or both. You get observations for each interview, findings, key insights with evidence strength, opportunity areas and what is still unknown, and what it all means for the decisions ahead.
Every insight can be traced back to the observations and interviews it rests on. The result is a knowledge base you can save and ask questions later.
- ObservationsWhat each respondent does, experiences or reacts to
- FindingsPatterns supported by several observations
- Key insightsWhat, why and why it matters, with evidence strength
- OpportunitiesWhere there is value to create, not finished solutions
- DecisionsWhat the results mean, and what is still unknown
What do you want to do?
I help you work through qualitative research you have done.
Already have a knowledge base? Open the file you saved to ask new questions or run a new synthesis.
That file is not a knowledge base from Research Synthesis.
Upload the material
Reading the files …
Paste text instead
Paste the whole text, at least a few paragraphs.
Nothing is counted or classified while you upload. When everything is here, the whole material is inventoried at once.
Analysis register
Every file is classified from its content, not its name. Check that this is the material you want to analyse, and correct a type if it is wrong.
A check that nothing is missing. Your number is never changed; the analysis waits until the register and the number match.
Finding the main questions in your interview guide …
Taken from your interview guide. Edit them if they are not right.
Add more context Optional
Used to say what the insights mean for what you develop, only where the interviews support it.
Part of each observation’s provenance. Left empty, it says “missing in the material”.
Anonymisation
For example teacher, customer or nurse. Their names are replaced with this word and the interviewer’s with “Interviewer”, in your browser, before anything is sent. Contact details are removed too, and the analysis never writes out names of other people mentioned: it uses their role.
Remove more names before sending Optional
If you do not want some names to leave your computer at all, for example a workplace or a town, add them here.
The material is sent to an AI service (Claude by Anthropic) for the analysis, with speakers’ names, email addresses, phone numbers, links and ID numbers removed first. Nothing is saved on my side: save your knowledge base before you leave the page. Two studies are free; after that you can sign up with your name and email to keep using the tool for free.
Observations
Evidence-near observations from each interview, with a coverage note. Review each card against its source before the observations are stored or used: approve, correct or reject it.
Continue to phase 2: cross-interview synthesis?
In the synthesis, the observations from all interviews are analysed together, to understand what the material shows as a whole: recurring patterns, important variation and differences, contradictions, relationships and mechanisms, findings, key insights, possible opportunity areas and remaining uncertainties. The analysis goes from single observations to a fuller understanding: observations → findings → key insights → opportunity areas. If you have decision material, the synthesis can also show what the results mean for the decisions ahead.
The analysis ends here. Save the observations as PDF and as a knowledge base file: you can open the file here later and run the synthesis without starting again from the transcripts.
How do you want the result structured?
Not sure what to choose? Help me choose
What the interviews show
Concept test
A concept test is part of the material. Do you also want the concepts analysed separately? Each concept gets an evidence class from A to D, and what the test supports and does not support, kept apart from the main synthesis.
Decisions ahead
What the material still does not support, what the results mean for the decisions ahead, and the recommended next step.
The decisions you are facing
Add decisions, assumptions or hypotheses, and see what the research means for each one. They are kept apart from the respondents’ evidence. Without them, you get general implications.
Ask your knowledge base
Write a decision you are facing or a question. The answer uses only this research, shows the evidence for and against, and says what is still unknown.
The analysis is done
The observations can already be saved on their own, from the observations step. Do you also want a PDF of the whole analysis? Choose “Save as PDF” in the print window. The knowledge base file holds everything, so you can open it here again later.
Finished. Remember that nothing is saved on my side: save the knowledge base file if you want to come back to the analysis.
Feedback
Help me make this tool really useful
How did it work for you, what value did it give you, and what would make it better? All feedback is welcome, big and small. I read everything.