How to generate a data document in DataLens
Documentation is the part everyone skips. Generating it from the model and pipeline you already built removes the excuse.
What you will do
You will produce a written document — an architecture document, a model specification, a mapping, a runbook or an analysis report — from the work already in your workspace.
Document types are organised by who needs them, because an architect and a business user need very different documents about the same system.
When this is useful
- A project needs an architecture document and nobody has time to write one.
- You are handing a system over and the knowledge is currently in your head.
- A stakeholder needs a written explanation rather than a screen.
- A review or approval requires documentation as an artefact.
Before you start
- The underlying work done — Documents are generated from your models, mappings, pipelines and analysis. There must be something to document.
- Knowing who it is for — The document types are grouped by persona — architect, business user, analyst, analytics engineer, engineer, scientist. Pick from the reader's side.
Steps
Open the document generation screen
Documents are generated from work already in your workspace, so build the model, mapping or analysis first.
The document type list, grouped by persona.
Choose the persona
Document types are organised by who reads them. Architect covers architecture documents, model specifications and handover documents. Business user covers executive briefs and insight summaries. Analyst, analytics engineer, engineer and scientist each have their own set.
Pick the document type
Available types include a Data Architecture Document, a Data Model Specification, a Technical Handover Document, a Pipeline Design Document, a Data Governance Summary, a Source-to-Target Mapping, a Pipeline Runbook, a Data Quality Summary, a Column Profiling Workbook, a Transformation Specification, an Executive Dashboard Brief, an Insight Summary Report, a Scenario Analysis Summary and others.
Each carries a status badge showing whether it is READY or PARTIAL, so you know what to expect before generating it.
The chosen document type with its status badge.
Check the status badge
READY means the document generates in full. PARTIAL means it generates with less coverage.
The badge is there rather than hidden precisely so you are not surprised by the output.
Generate it
Generate the document. It is built from the models, mappings, pipelines and profiling in your workspace.
The generated document.
Review before circulating
Read it. A generated document reflects what is in the workspace — if the model is incomplete, the document will be too, faithfully.
The generator saves you the writing, not the reviewing.
What happens next
A document reflects the workspace at the moment it was generated. When the model changes materially, regenerate rather than editing the old one.
Documents pair naturally with model exports — a written specification alongside an SVG diagram covers most audiences.
Example
A team finishes a source model and mapping for a migration. Rather than spending a week writing, they generate a Data Architecture Document and a Source-to-Target Mapping, review both, and take them to the design review — which becomes a discussion of the design rather than a discussion of whether it was written down.
Tips
- Do the modelling first. A document generated from a thin workspace is a thin document.
- Read the status badge before generating — PARTIAL tells you what to expect.
- Choose the type by reader, not by title. An executive brief and an architecture document describe the same system very differently.
- Regenerate after material changes rather than editing yesterday's file.
Limitations
- Documents are generated from what is in your workspace. They cannot describe work that has not been done.
- Types marked PARTIAL generate with less coverage than those marked READY.
- A generated document is a strong draft, not a signed-off deliverable. Review it before circulating.
- A document is a snapshot. It does not update when the underlying model changes.
Related how-to guides
Explore a source model
See the tables, columns and relationships of a system you did not design — as a diagram you can read.
Map source to target
Connect source columns to target columns, see what is unmapped, and export a source-to-target mapping.
Trace where a column came from
Follow a field back from the dataset you are looking at to the source it came from, step by step.
Publish and export data
Export as CSV, TSV, JSON, JSONL or Excel, publish an API endpoint, or generate a PDF or Word report.
Related questions
More of these on the DataLens FAQ page.
Can DataLens generate data architecture documentation?
Yes. Document types include a Data Architecture Document, Data Model Specification, Technical Handover Document, Pipeline Design Document, Source-to-Target Mapping, Pipeline Runbook and Data Governance Summary, among others — generated from the models, mappings and pipelines already in your workspace.
Are generated documents ready to send?
Treat them as strong drafts. They are generated from what is in the workspace, so an incomplete model produces an incomplete document — faithfully. Each type also carries a READY or PARTIAL badge so you know the coverage before generating.
Can I get documentation aimed at business readers rather than architects?
Yes. Document types are grouped by persona — architect, business user, analyst, analytics engineer, engineer and scientist — with types such as an Executive Dashboard Brief and an Insight Summary Report for non-technical readers.
Try this in DataLens
DataLens is in private beta. Request access and work through this guide on your own data.
Request beta access