Written by our assessment team: what the criteria mean in practice, the evidence that works, and where learners get caught out.
What this unit is really about
Unit 523 carries 10 credits — more than any other unit in either qualification — which tells you the expected volume of work is substantial. It is worth choosing deliberately: on a Certificate or Diploma it can cover a large share of the credit requirement in one unit, but it is a poor choice for anyone without access to real organisational data.
The shape is: understand data's role (six criteria), source and evaluate data (three), then use it (three). The practical half needs internal and external data, an evaluation of its quality, an analysis of its strengths and limitations, real analysis, presentation using visualisation tools, and a justification of the process you used.
The evidence that works
| Learning outcome | Evidence that works well |
|---|
| LO1 — role of data | The characteristics of a data-driven culture, legal and ethical responsibilities in data management, an assessment of the benefits, how data supports operational and strategic decisions, how presentation supports decisions, and the impacts of inaccurate or misinterpreted data |
| LO2 — source and access | Internal and external data actually sourced, an evaluation of its quality, and an analysis of its strengths and limitations for your purpose |
| LO3 — use data | Analysis with the methods stated, charts or dashboards produced with visualisation tools, and a justification of the process |
Criterion 2.1 requires external data, which learners often skip. Free and credible UK sources: ONS statistics, Nomis labour market data, sector regulator publications, trade body benchmarks, Companies House filings for competitors, and local authority open data.
Data quality, evaluated rather than assumed
Criterion 2.2 asks you to evaluate the quality of the data you accessed, so use explicit dimensions: accuracy, completeness, timeliness, consistency, validity, uniqueness and relevance. Then apply them honestly — most real workplace data fails at least two. Duplicate customer records, a field that became optional in 2023, timestamps recorded in local time on some systems and UTC on others, a "reason code" that 60% of users leave as the default.
Reporting those flaws is not a weakness in your evidence; it is criterion 2.3. And it protects criterion 3.3, because your justification of the process should say what you did about them — excluding a period, cleaning duplicates, treating a field as indicative only.
Legal and ethical, precisely
Criterion 1.2 needs accuracy. For Great Britain that is the UK GDPR and the Data Protection Act 2018: a lawful basis for processing, purpose limitation, data minimisation, accuracy, storage limitation, security, and accountability. Add the points that bite in practice: special category data needs an additional condition, data subject access requests apply to anything you hold about identifiable staff, and using data collected for one purpose to make a different decision is exactly what purpose limitation restricts.
The ethical half goes beyond the law: whether people would expect their data to be used this way, proxy discrimination where a neutral variable stands in for a protected characteristic, the surveillance risk in productivity monitoring, and transparency where analysis affects individuals. If you use AI or automated tools, note your organisation's policy and the accountability question — the manager remains responsible for the decision.
Visualisation and the misinterpretation criterion
Criterion 3.2 requires visualisation tools — a spreadsheet's charting is acceptable, as are Power BI, Tableau or Looker Studio if you have them. Criterion 1.5 and 1.6 then pull in opposite directions and should be answered together: presentation aids decisions, and presentation misleads. Truncated axes, dual axes implying correlation, percentages on tiny bases, and averages hiding bimodal distributions are all worth naming — ideally with a real example of a chart in your organisation that misled someone, redrawn correctly.
Useful reading
