From Report Production to Evidence Assurance: A Practical Review Method for Council Reporting

From report production to evidence assurance

Council reporting is often discussed as a question of templates, dashboards and deadlines. Those matter, but they are not the core problem. The deeper question is whether a reader can follow a reported result back to a clearly defined measure, a reliable source, a documented calculation, a responsible owner and a decision that the information is meant to support.

When that chain is unclear, a polished report can still be difficult to trust or act on. Staff may spend time reconciling versions, chasing evidence and explaining numbers after the reporting period has ended. The objective should not be to eliminate spreadsheets at any cost. It should be to make the reporting process understandable, controlled and proportionate to the decisions and risks involved.

This article sets out a practical review method for one council reporting process. It is a method guide, not a case study or a claim about any particular council’s systems or performance.

Why council reporting has to connect several layers

Council reporting sits across strategic commitments, services, work programs, budgets, data and public accountability. The architecture differs by state and territory, and a reporting review must begin with the applicable jurisdictional and council requirements—not assume a single national template.

For example, New South Wales’ Integrated Planning and Reporting framework links a Community Strategic Plan to a four-year Delivery Program, an annual Operational Plan and a resourcing strategy. The NSW Office of Local Government describes the Operational Plan as identifying the projects and activities for the year, with the annual budget and revenue policy, and says the relationship to the Delivery Program and Community Strategic Plan should be clear. Victoria uses a different model: its Local Government Performance Reporting Framework is mandatory for Victorian councils and includes 59 service-area measures and a 27-item Governance and Management checklist. These are jurisdiction-specific examples, not requirements that should be applied unchanged across Australia. (NSW Office of Local Government (https://www.olg.nsw.gov.au/councils/policy-and-legislation/integrated-planning-and-reporting); Local Government Victoria (https://www.localgovernment.vic.gov.au/strengthening-councils/performance-reporting))

The practical implication is that an indicator rarely stands alone. A sound review asks how the measure relates to the council’s purpose, plan, funded activity, operating data and reporting obligation. If staff cannot explain that relationship in plain language, more dashboarding will not fix it.

Start with the decision, not the spreadsheet

Before tracing data, ask what decision or accountability question the report is meant to serve. Is it helping councillors monitor delivery of a commitment? Helping managers redirect resources? Showing residents what has changed? Demonstrating compliance with a prescribed reporting framework? Different uses may require different levels of detail, timing and assurance.

A useful test is: If this number changed materially, who would need to know, what would they ask, and what action could follow? If no one can answer, the measure may be reported because it is easy to count rather than because it helps explain performance.

My view is that councils should be wary of “dashboard first” projects. A visual interface can make weak definitions look authoritative. Define the decision and the measure first; choose the display and technology second.

The evidence chain: eight questions for a reporting review

A focused review can follow a single indicator or report item through eight linked questions. The goal is to expose gaps and unnecessary effort, not to create documentation for its own sake.

1. What is being measured, and why?

Record the measure’s purpose, population or service boundary, reporting period, unit, direction of improvement and relationship to the relevant plan or obligation. State whether it measures an input, activity, output, service quality, outcome or impact. A count of completed inspections, for example, is not automatically evidence that risk has fallen; it may be an output that needs to be read alongside a quality or outcome measure.

The Australian National Audit Office’s guidance on meaningful performance information (https://www.anao.gov.au/work/insights/reporting-meaningful-performance-information) says measures should relate directly to purpose, be clear and logical, be complete and balanced, remain contemporary, and have an owner. Although the ANAO guidance is for Commonwealth entities, these are useful design questions for councils when applied with the council’s own legislative and policy framework in view.

2. Is the definition precise enough to reproduce?

Capture the numerator, denominator, exclusions, cut-off date, treatment of reopened or cancelled cases, and any business rules. Avoid words such as “resolved”, “on time” or “completed” unless the process defines them. If teams interpret a term differently, the reported result may be a mixture of local conventions rather than a comparable measure.

A measure definition should be brief enough to use, but specific enough that a trained person who did not prepare the report can understand what is included. A definition that exists only in one analyst’s memory is a continuity risk.

3. Where does the evidence originate?

Map each value back to its source: a line-of-business system, finance system, asset register, approved survey, case-management record, manual log, or a third-party return. Note who can access the source, how often it is updated, whether it is authoritative for this purpose, and what happens when a value is corrected.

A spreadsheet can be a perfectly reasonable reporting tool. It becomes fragile when there are multiple “final” copies, hidden transformations, undocumented manual edits, unclear ownership or no record of which version fed the published report. The tool is not the control; the repeatable and reviewable process around it is.

4. Can another person reproduce the calculation?

Document the steps from source extract to reported figure: filters, joins, formulas, manual adjustments, aggregation, rounding and validation. For a manually entered value, record who supplied it, when, under what definition and who checked it. Where another organisation supplies data, record the agreed method and the checks performed before relying on it.

The ANAO recommends that organisations document how measures are calculated, the data source and collection frequency, and relevant business rules and assumptions. This does not mean every council needs a large technical specification. It means the amount of explanation should be enough for the measure’s importance and risk.

5. Who owns, prepares, checks and approves it?

Separate four practical roles where the process warrants it: the business owner accountable for the meaning and use of the measure; the data steward or source-system owner; the person compiling the result; and the reviewer or approver. In a small team, one person may hold more than one role, but the overlap should be visible and compensating checks considered.

“Owned by the reporting team” is usually too vague. Name a role or position, define what it is responsible for, and provide a handover route when staff change. The owner should be able to explain both the result and its limitations.

6. What exceptions or changes need explaining?

A number without context can mislead. Identify missing periods, changed definitions, system migrations, exceptional events, policy changes, unusual weather, contractor changes or other factors that affect interpretation. Keep a concise exception log, and distinguish a real change in service performance from a change in how it was counted.

A target miss is not automatically a failure of management, just as a target met is not automatic proof of success. The report should explain what is known, what is uncertain, what was within the council’s control and what action follows. Balanced reporting is more credible than a page that only celebrates favourable results.

7. Is the supporting record retrievable and appropriately handled?

For material reported results, identify the records needed to explain the figure and make sure they can be located by someone other than the preparer. Define the appropriate recordkeeping location, version and retention approach under the organisation’s records policy. Do not create unnecessary duplicate stores of personal, sensitive or confidential information merely to make a reporting pack feel complete.

Evidence assurance is not “collect everything”. It is being able to retrieve the right evidence, understand its provenance, and demonstrate why it supports the reported claim—while respecting privacy, security and records obligations.

8. Does the report help the reader understand and act?

Use plain language, stable definitions, appropriate comparisons and a brief explanation of material variation. Distinguish reported activity from outcomes. State limits where they matter. Use charts only when they clarify a comparison or trend; a chart that hides a denominator or changes its scale without explanation can reduce, rather than improve, understanding.

The ANAO advises that performance information should be fair, balanced and understandable, with context, consistent terminology and transparent discussion of results above or below expectations. It also cautions against unnecessary information: the aim is a meaningful account, not the largest possible data pack.

A worked example—illustrative only

Consider a hypothetical quarterly measure called “percentage of customer requests completed within service standard”. This is not a C51 client example. Before reporting it, a council might need to answer:

• Does the denominator include all requests received, only eligible requests, or only requests closed in the quarter?

• How are requests that are withdrawn, duplicated, transferred or reopened treated?

• Is the service standard measured in calendar days or business days, and does the clock pause while the council waits for information?

• Which system timestamp is authoritative, and what happens when a record is corrected?

• Are requests from different service areas being combined even though they have different standards or complexity?

• Who checks the extract and investigates a surprising change?

• Does the measure show timeliness only, or is it paired with a quality, repeat-contact or customer-experience measure?

There is no universal answer to these design choices. The important point is to make the choices explicit, approved by the accountable business owner, and applied consistently. If a definition changes, the report should state when and why, and consider whether trend comparisons remain valid.

A practical maturity lens

The following five stages are a diagnostic lens for discussion, not a formal standard or an external certification scheme:

1. Person-dependent: results rely on individual memory, manual chasing and local files. Turnaround and confidence vary with staff availability.

2. Repeatable: a recurring timetable and basic template exist, but definitions, checks or source records may be inconsistent.

3. Defined: measure definitions, roles, data sources, calculation rules and approval steps are documented for the reporting process.

4. Traceable: material results can be followed back to evidence; changes and exceptions are logged; reviewers can reproduce key calculations.

5. Decision-integrated: reporting is timely enough to inform decisions; trends and limitations are interpreted; learning changes plans, service design or resource choices.

Most organisations will sit at different stages for different measures. That is normal. The aim should not be to “mature” every indicator to stage five. Focus first on measures that are material to public accountability, high-risk services, significant funding or decisions that are difficult to reverse.

A proportionate review in ten working days

A small review can produce useful clarity without turning into a transformation program. For one defined report or a handful of priority measures:

Days 1–2: agree the boundary. Confirm the reporting purpose, users, applicable jurisdictional requirements, period, measures in scope and decision owner. Ask which result is most difficult to explain or prepare.

Days 3–4: walk the process. Speak with the people who supply, compile, check and approve the information. Observe the real workflow and examine the actual source files and hand-offs—not only the documented process.

Days 5–6: trace selected measures. Follow a small sample from reported figure to source record and back through the calculation. Test one ordinary case and, where possible, an exception or correction. Do not infer that a sample proves the entire population is accurate.

Days 7–8: identify causes and risks. Separate definition issues, process friction, data-quality limitations, access or system constraints, skills gaps and governance gaps. Note which problems are factual findings and which are hypotheses that need validation.

Days 9–10: agree practical actions. Prioritise changes by public-accountability risk, decision value, effort and dependency. Assign an owner and due date. Define how the council will know the change helped—for example, fewer reconciliation steps, faster evidence retrieval, fewer unresolved exceptions or clearer interpretation by intended readers. Do not claim the improvement until it has been measured.

A useful output is a short measure register, a process/evidence map, an issue and risk log, and a prioritised improvement plan. The format matters less than clarity, ownership and follow-through.

Where automation helps—and where it does not

Automating a stable, well-defined transfer or validation can reduce repetitive handling and make a process easier to repeat. Automating an ambiguous definition can simply make the ambiguity faster and harder to see. Before automating, test whether the source is reliable, the rule is understood, exceptions are manageable and an accountable person can review the result.

A practical order is: clarify the measure; stabilise the process; agree the control; then automate the repeatable step if the benefit justifies the cost and risk. Keep a route for correction, explain changes to definitions, and avoid building a critical reporting process around a tool that no one can support.

Common traps—and a few firm opinions

More indicators do not automatically mean more accountability. A small, balanced set that answers the key questions is usually more useful than a long list of measures no one interprets. The ANAO similarly notes that measures should collectively provide a meaningful and balanced picture; a single measure is rarely enough to explain a complex purpose.

A target is not a strategy. Targets can focus attention, but a target with a weak definition or no explanation of external factors can encourage the wrong behaviour. Use targets alongside context and, where appropriate, quality or outcome information.

A dashboard is not a control framework. Visualisation can help users see patterns. It does not, by itself, establish source authority, calculation integrity, approval responsibility or evidence retention.

Avoid treating “spreadsheet” as the root cause. Replacing a spreadsheet with a new platform may be justified, but only after understanding the process. Otherwise, the organisation risks digitising unclear ownership and adding a second system to reconcile.

Do not confuse assurance with bureaucracy. Controls should be proportionate. The right question is not “How many documents can we attach?” but “What would a reasonable reader need to verify this material result, and can we retrieve it without rebuilding the story each reporting cycle?”

What good looks like

A reporting process is in better shape when a new staff member can understand the measure; the owner can explain why it matters; the preparer can reproduce the figure; the reviewer can see what changed; and the reader can understand what the result does and does not mean. The process should support decisions and public accountability without creating an unnecessary evidence burden.

For a council reviewing its own reporting, a useful first step is to choose one important measure and trace it end to end. Ask where the definition lives, who owns it, what source supports it, how the result is calculated, what exceptions occurred and what decision the report informs. The answers will show whether the main opportunity is clearer definitions, better hand-offs, stronger evidence records, more useful interpretation—or, only then, a technology change.

The objective is not a perfect reporting system. It is a reporting process that people can explain, reproduce and improve.

References

• NSW Office of Local Government — Integrated Planning and Reporting (https://www.olg.nsw.gov.au/councils/policy-and-legislation/integrated-planning-and-reporting)

• Local Government Victoria — Performance Reporting (https://www.localgovernment.vic.gov.au/strengthening-councils/performance-reporting)

• Australian National Audit Office — Reporting Meaningful Performance Information (https://www.anao.gov.au/work/insights/reporting-meaningful-performance-information)

• Australian National Audit Office — Performance Measurement and Monitoring (https://www.anao.gov.au/work/insights/performance-measurement-and-monitoring-developing-performance-measures-and-tracking-progress)