Popular
Start here Quickstart Sign up, import a dataset, and build your first cross-tabulation. Five minutes, end to end. Everything beyond this is optional until you need it.Building analysis Build tables and analysis Drag variables to rows and columns; the cross-tabulation computes instantly. Layer on measures, weights, significance tests, and expressions to refine it — and use formula cells for bespoke calculations.Working with the agent Agent recipes Patterns for getting reliable, high-quality work out of the agent — from a one-shot dataset description to a full draft report. Each recipe shows the prompt, what the agent does, and how to verify the output.Building analysis Methodology How Recense computes the numbers — significance tests, weighting, missing-value treatment, and the reasoning behind each default. Each section names the test, when it runs, and the published reference behind it.Account & reference Security and privacy Recense is local-first by default — parsing, tabulation, and analysis run in your browser. Cloud projects and cloud datasets are encrypted on your machine before upload, and enterprise hosted compute only applies to datasets your organisation explicitly publishes.
Start here
A five-minute path from sign-up to your first cross-tabulation.
Working with data
Import, explore, reshape, and save your survey data.
- Import and review Drop an SPSS .sav or CSV file onto Recense to parse it locally — schema, value labels, weights, and groups load directly. Your data never leaves your browser.
- Explore datasets Use the Dataset Explorer to verify variables, fix labels, and confirm group structure before you tabulate. Three views — Variables, Question Groups, Raw Data — share the same source of truth.
- Stack question groups Reshape a multi-item grid so each item becomes its own row. Stacking lets you analyse the items as a single variable instead of comparing N parallel columns.
- Code open-ended responses Turn free-text answers into a categorical variable in six steps: pick the question, generate themes, review them, classify responses, review the classifications, and publish. Recense uses AI for the heavy lifting and gives you full manual override.
- Save and manage projects Autosave, local export, and encrypted cloud save — three save paths that capture your dataset, canvas, instructions, and agent history together. Pick the one that matches how portable and how shared you need the project to be.
Building analysis
Tabulate, chart, and apply statistical methods.
- Build tables and analysis Drag variables to rows and columns; the cross-tabulation computes instantly. Layer on measures, weights, significance tests, and expressions to refine it — and use formula cells for bespoke calculations.
- Chart and share results Build a chart from any table — it stays linked to its source. Present the canvas in a clean walkthrough, or publish a password-protected snapshot for stakeholders.
- Methodology How Recense computes the numbers — significance tests, weighting, missing-value treatment, and the reasoning behind each default. Each section names the test, when it runs, and the published reference behind it.
Working with the agent
Set up the agent and collaborate with it on real analysis.
- Set up the agent The agent has the same capabilities whichever path you choose — what differs is where the model runs and how you authenticate. Pick BYOK for provider control, built-in for zero setup, or MCP to drive Recense from Claude or ChatGPT.
- Agent recipes Patterns for getting reliable, high-quality work out of the agent — from a one-shot dataset description to a full draft report. Each recipe shows the prompt, what the agent does, and how to verify the output.
- Bring your own keys (BYOK) Add your own provider API key in Settings → Provider keys and the agent calls that provider directly from your browser. Keys stay on your device unless you choose to save them, end-to-end encrypted, to your account.
- Use the built-in agent The built-in agent routes requests through Recense — no API key required. Activate it once in Settings → Hosted agent; usage is metered against your plan's monthly allowance and shown in Settings → Usage.
- Connect MCP clients Connect Claude Code, Claude Desktop, or ChatGPT to your live Recense session in four steps: load a dataset, generate an API key, add the relay endpoint, and test with a read-only prompt. The assistant works with your live session — not a separate copy.
- Give the agent context Tell the agent what the survey is, how you like to work, and which documents it can draw on — once, on the Instructions tab, instead of in every prompt. Context lives with the project.
Account & reference
Plans, security, shortcuts, and troubleshooting.
- Team and billing Access, plan allowances, account controls, and team workspaces. During private beta, availability depends on your invitation.
- Security and privacy Recense is local-first by default — parsing, tabulation, and analysis run in your browser. Cloud projects and cloud datasets are encrypted on your machine before upload, and enterprise hosted compute only applies to datasets your organisation explicitly publishes.
- Keyboard shortcuts Quick reference for canvas, table editing, and general app shortcuts. On Windows and Linux, use Ctrl wherever this page says Cmd. Press ? on the canvas to see the same list in the app.
- Troubleshooting Quick answers to the most common problems — import, data, agent, MCP, and performance. If your issue isn't here, use the in-app Support panel to send a report.