Every Dataset in Qualibr moves through the same loop. AI drafts the work at each stage — proposing structure, Codes, and candidate evidence — and none of it becomes part of your analysis until you review it. This page walks through what actually happens at each step.
Create a Project and configure the first Dataset: method, coding approach, context, and languages.
This is the only stage where you make research-design decisions — which method fits the data, whether coding is inductive, deductive, or hybrid, and what context Qualibr should keep in view. Everything downstream inherits these choices; nothing here is guessed by AI.
Import text-based research material, check what is ready, and review analysis scope and cost before starting.
Interview transcripts, focus-group transcripts, open survey responses, and research documents go in as-is — Qualibr extracts and structures the text so you are not the one reformatting files. Before you run anything, you see the credit estimate for that Source, not after.
Work with Codes and the method-specific structure. Accept, edit, reject, merge, or revisit work as needed.
Qualibr proposes Codes and links each one to its exact source Excerpt as it works through your material. Nothing here is a one-shot output: accept a Code, rename it, merge two into one, reject it outright, or send a batch back for another pass — at any point, not only at the end.
Review evidence, limitations, and interpretation; export a Dataset report or Codebook from the relevant workspace.
A Finding stays a draft until you review its supporting, contradicting, and limiting evidence and decide what it actually means. Once you are satisfied, export a Dataset report or Codebook straight from the workspace — the same evidence trail you built during review goes with it.
The route is deliberate, not necessarily linear or irreversible — you can revisit earlier stages from inside the Dataset workspace at any point.
20 free credits, no card required — enough to run this loop on a real interview before deciding whether to pay for anything.