
Step 6 - Sustain, improve and connect |
Embed data inventory as a living capability within the organisation. A data inventory that is not actively maintained loses its value quickly. This step is not a final phase but the ongoing operating cycle of your data management capability and maturity.
The goal is to make inventory maintenance a normal part of how your organisation works, not a special exercise that happens when someone has time. Practical mechanisms include:
At least annually, your organisation should review the inventory against a set of improvement questions:
Document your review cycle as a standing agenda item in your data governance committee. Record what changed, why and what will be prioritised in the next cycle. This creates an audit trail and demonstrates continuous improvement to auditors and oversight bodies.
A data inventory becomes significantly more valuable when it connects to the broader data landscape, inside and outside the organisation.
Internally
Externally
A data inventory tells you what data you have. Data quality practice tells you how good it is. These two capabilities reinforce each other — a mature inventory includes quality metadata; a quality program depends on knowing what data assets exist.
The ABS Data Quality Framework provides a practical, widely recognised model for assessing and communicating data quality. Its seven dimensions can be applied to individual data assets in your inventory:
| Quality dimension | What it asks | How to apply in your data inventory |
| Institutional environment | Is there a sound, stable governancestructure for this data? | Record custodian, owner and governance arrangements |
| Relevance | Does this data meet the needs of current and likely users? | Note intended use cases; flag if relevance is declining |
| Timeliness | Is the data current enough to be useful? | Record update frequency and date lasty updated |
| Accuracy |
Does the data correctly represent what it measures? |
Note known accuracy limitations or data quality issues |
| Coherence | Is this data consistent with related assets over time? | Note dependencies and known consistency issues |
| Interpretability | Can users understand and correctly use this data? | Include description, field definitions, and caveats |
| Accessibility | Can authorised users find and access this data easily? | Record access level, location, and request process |
You don't need to formally rate every asset against all seven dimensions from day one. Start by capturing timeliness (update frequency and last modified date) and accessibility (access level and location).