Skip to main content

Search the Commons

Type at least 2 letters. Use the arrow keys to move through results, Enter to open, Escape to close.

Ask a question instead: answers drawn from across the Commons, with links to the source. Or browse the site map.

Data quality at a glance

Pictures instead of tables. See who does each part of data quality work, how much it matters, how the steps follow on, how the records connect, and how a rule travels from your desk to the National Data Library.

In short: a plain English summary
  • This page uses pictures to explain data quality work.
  • It shows who does each step.
  • It shows which steps matter most.
  • It shows how a rule moves from your computer to other places.
  • Every picture can also be read as a list.

Choose a workflow

The first two pictures show the workflow you choose. They come from the Commons capability model, the same source as the capability and data models page.

Who does what, and how much it matters

Each step of the workflow is a point on the map, in the style of a Morgan Map. Left to right: who does the work, from people to the toolkit. Bottom to top: how much it matters to trust, from good practice to required by the Data Asset Management policy. Lines show what leads to what. Choose a point to see its details.

Show the map as a list

Follow the steps

The same workflow from start to finish. Each row is a role; dashed lines loop back.

Wide picture: scroll sideways to see it all, or open the list below.

Show the steps as a list

How records connect

The records a data quality service keeps, grouped by area. An arrow means one record points to another, for example an issue points to the rule result that raised it.

Wide picture: scroll sideways to see it all, or open the list below.

Show the records as a list

How a rule travels

Lineage of your data quality rules through the Commons: what each step produces and where it ends up. Every box is a real feature you can open.

Wide picture: scroll sideways to see it all, or open the list below.

Show the journey as a list

How each step is placed

Nothing on the map is a guess. Each step's position comes from the capability model by these rules, so it changes when the model does.

Who does it (left to right)

  • People: a task done by a person, with no Commons tool.
  • Decision: a yes or no choice made by a person.
  • People with a tool: a person's task that a Commons tool supports.
  • The toolkit: a task the toolkit does automatically.

How much it matters (bottom to top)

  • Required: the step cites the Data Asset Management policy or its mandatory guidance on critical data assets, metadata and ownership.
  • Recommended: the step cites other published guidance, such as the action plan guide, issues framework or Government Data Quality Framework.
  • Good practice: no guidance cited.
  • A decision takes the level of the most important step it leads to.

Reading the corners

Top left: high stakes and done by people, so keep human judgement and make it easy to evidence. Top right: high stakes and automated, so automate with checks and a person to review. Bottom right: let the toolkit do it. Bottom left: simplify with templates.

Inspired by the Morgan Map, which maps the decisions, evidence and execution in a service by automation and criticality to trust. Each step's details say which of the three it is: a decision, evidence (it produces a record) or execution.