What Should a Multi-Site Health and Safety Dashboard Show?

A multi-site health and safety dashboard should show more than accident totals. It should help a manager decide where attention is needed now: which serious risks are not adequately controlled, which corrective actions are overdue, which assessments or checks are missing, where incidents or near misses are repeating, who owns the next step and whether the underlying data is complete enough to trust.

What decision should the dashboard support?

Start with the decisions people must make, not the charts available in the software. A site manager may need to know what must be made safe before the next shift. A regional manager may need to decide which location needs help. A senior leader may need to see whether the same control is failing across the organisation. The dashboard should make each of those decisions easier.

A practical test is to ask whether a user can answer five questions without assembling a separate spreadsheet:

• What is outside the agreed standard or deadline?
• Which exception could cause the most serious harm?
• Who owns the next action, and when is it due?
• Is this isolated, or is the same issue appearing elsewhere?
• Can the user open the assessment, investigation, check or evidence behind the number?

This is why an exception-led view is usually more useful than a page of totals. Ten completed low-risk actions should not hide one overdue action on an uncontrolled high-risk hazard.

Is a health and safety dashboard legally required?

A particular dashboard format is not a general legal requirement in Ireland, Great Britain or the United States. The underlying duties and records still matter.

In Ireland, the Safety, Health and Welfare at Work Act 2005 requires employers to identify hazards, assess risks and prepare a written safety statement. HSA guidance describes active and reactive monitoring, review, audit and corrective action as parts of an effective safety and health management system. A dashboard can support that work, but it does not replace a suitable risk assessment, safety statement, consultation, competent advice or action at the workplace.

In Great Britain, employers must manage health and safety risks under the Health and Safety at Work etc. Act 1974 and the Management of Health and Safety at Work Regulations 1999. HSE's leadership guidance recommends that boards receive preventive information as well as incident data and understand what is happening on the ground. That is guidance on effective management, not a prescribed dashboard specification.

In the United States, OSHA's Recommended Practices for Safety and Health Programs promote proactive hazard finding, measurable goals, assigned responsibilities and performance monitoring. They are recommended practices. They do not replace the Occupational Safety and Health Act, applicable OSHA standards, recordkeeping rules or state-plan requirements.

For a multi-jurisdiction organisation, keep the legal registers, reporting thresholds and required records for each location separate. The dashboard can provide a common management view, but it should not imply that one country's rules apply everywhere.

1. Corrective actions and serious exceptions

Corrective actions are often the most useful opening view because they connect a known problem to a responsible person and deadline. Show:

• open and overdue actions, separated by risk or priority;
• actions arising from serious incidents, high-risk assessments, regulatory visits or failed critical controls;
• owner, due date, age and site;
• interim controls where the final fix is not yet complete;
• actions awaiting verification, not just those marked complete;
• repeat extensions or reopened actions; and
• common themes across sites, contractors or work activities.

Separate completed from verified effective. Uploading a photograph or ticking a box may show that work was attempted; it does not always show that the control now works. Define who can verify closure and what evidence is proportionate to the risk.

Age matters as well as count. A location with four new low-priority actions may need less attention than one with a single high-priority action that has been overdue for six weeks.

2. Risk-assessment coverage and change

A coverage percentage can be useful, but only when the denominator is clear. “95% complete” means little unless leaders know whether the inventory includes all relevant sites, tasks, equipment, substances and groups of workers.

Useful risk indicators include:

• assessments due or overdue for review;
• new activities or sites not yet assessed;
• high residual risks and the controls still required;
• assessments affected by an incident, change, new equipment or new information;
• temporary controls that have remained in place beyond their planned period;
• worker consultation or approval still outstanding; and
• assessments with actions that have not been closed and verified.

A review-date flag is only a prompt. An assessment may need review before that date when work, people, equipment, substances or evidence changes. Conversely, a recently reviewed document is not evidence that controls are present in practice. Link the assessment view to inspections, incidents and actions that test it.

3. Control checks, inspections and audits

HSA distinguishes active monitoring before things go wrong from reactive monitoring after an incident or failure. HSE similarly recommends preventive information alongside accident and ill-health data. The dashboard should therefore show whether important controls are being checked and whether those checks find meaningful issues.

Choose indicators around the controls that matter most to the organisation. Examples include statutory examinations, equipment inspections, fire precautions, workplace transport checks, contractor controls, exposure monitoring, lone-worker arrangements or medication-handling checks in care settings. The exact set depends on the risk profile.

For each programme, show more than the number of forms submitted:

• checks planned, completed and missed;
• critical failures or failed control questions;
• findings still open and their age;
• repeat findings after a previous close-out;
• sites with suspiciously identical or consistently perfect results;
• audit recommendations and whether implementation was tracked; and
• evidence that an action changed the condition in the workplace.

A rising number of reported hazards may be a positive sign if workers are engaged and hazards are being fixed. Interpret the measure with close-out speed, severity and repeat findings rather than treating every increase as deterioration.

4. Incidents, near misses and recurring causes

Lagging indicators still matter. Show injuries, work-related ill health, dangerous events, property damage and near misses at an appropriate level of detail. Where rates are used, state the denominator and period. Do not compare a small shop with a large distribution centre using raw counts alone.

Useful views include:

• incident and near-miss trends by site, activity and event type;
• actual and potential severity;
• investigations awaiting completion;
• recurring immediate and underlying causes;
• events linked to the same equipment, contractor, shift or control;
• statutory reports due or submitted under the rules for that jurisdiction; and
• actions created, closed and verified following investigation.

Do not use a low incident count as proof that a site is safe. A quiet site may be controlling risk well, may be under-reporting or may simply not have experienced the consequence yet. Compare incident data with inspections, risk exposure, worker reports and action completion.

Protect privacy. Senior dashboards rarely need names or medical detail. Limit access to personal information and display only what the user's role requires.

5. Training and competence gaps

Training completion can be a leading indicator, but attendance alone is not competence. Show the gaps that affect current work, such as mandatory induction, task-specific instruction, authorisation, refresher dates and supervised sign-off.

Where possible, connect the indicator to the risk or role it supports. “Twelve courses overdue” is less actionable than “two forklift operators at Site A have expired authorisation and are rostered this week.” Keep sensitive employment information behind appropriate permissions.

Consider whether the organisation also needs evidence of briefings following a changed risk assessment, incident learning or new control. A dashboard should help managers see who still needs the information, without implying that a completion record proves the person will perform the task safely.

6. Reporting and data quality

A dashboard can look precise while the underlying records are incomplete. Include a small set of data-quality checks so users know what they are looking at. These may include:

• sites that have not submitted expected checks or reports;
• records with no owner, due date, risk rating or location;
• duplicate or inconsistent site and hazard names;
• actions closed without the required evidence or verification;
• assessments with no recent consultation or review evidence;
• large differences in reporting levels between comparable sites; and
• the date and time each view was last updated.

Use these indicators to improve the system, not to punish reporting. If managers believe that more near-miss reports will make their site look worse, the dashboard can suppress the very information leaders need.

How should sites be compared fairly?

Start with like-for-like groups: similar activities, staffing, opening hours and exposures. For incident rates, use an appropriate denominator such as hours worked, workforce size or transactions, and state it beside the figure. Very small denominators can produce volatile rates, so show the underlying numbers as well.

Do not create a league table from a single blended score. Weighting severity, compliance, reports, checks and training into one number hides assumptions and can encourage gaming. If a summary status is needed, publish the rules and allow users to open the components.

Trend each site against its own history as well as its peers. A site may still be better than average while deteriorating quickly. Another may appear worse because it has built a stronger reporting culture and is finding hazards earlier.

What should the dashboard layout look like?

A clear multi-site dashboard can use three levels:

1. Organisation view: serious exceptions, overdue high-priority actions, emerging themes and sites needing attention.
2. Site or region view: risk, control, incident, competence and completion trends for the selected location.
3. Record view: the assessment, finding, investigation, action, owner, evidence and change history behind the indicator.

At the top, show a short “needs attention” list rather than every available chart. Use consistent definitions and date ranges. Make risk and overdue status distinguishable without relying on colour alone. Display the reporting period and last refresh. Provide filters for site, region, work activity, contractor and risk theme only where the underlying data supports them.

Every indicator needs a named owner, definition, source and response rule. For example: if a high-priority corrective action becomes overdue, who is notified, when is it escalated, and who can accept a revised due date? A red tile without an agreed response is decoration, not control.

How often should the dashboard be reviewed?

Match review frequency to the decision. Site-level critical exceptions may need attention during a shift. Regional action and reporting gaps may be reviewed weekly. Senior leaders may review trends monthly or quarterly, with immediate escalation for a major incident, regulatory intervention or failure of a critical control.

HSE says systems should allow boards to receive both routine and incident-led reports. That means the calendar review must not become a reason to hold urgent information. Define the events that bypass the normal reporting cycle.

Periodically review the dashboard itself. Retire measures that no longer drive a decision, add indicators when the risk profile changes and sample the underlying records to confirm that the view remains reliable.

Common health and safety dashboard mistakes

• Showing only injury rates: this looks backwards and can hide uncontrolled risk or under-reporting.
• Counting activity instead of quality: one hundred inspections are not valuable if they miss the real hazards or findings stay open.
• Celebrating closure without verification: completion should not automatically mean the control is effective.
• Using unclear denominators: raw counts can make large or busy sites look worse without context.
• Comparing unlike sites: exposure, workforce and work type matter.
• Creating incentives not to report: reporting volume can reflect trust and engagement.
• Mixing legal regimes: a shared view must preserve jurisdiction-specific duties and reporting rules.
• Hiding the record behind the chart: users need to drill into evidence and ownership.
• Collecting everything: too many metrics dilute attention and increase data-entry burden.

A practical setup sequence

1. List the decisions required at site, regional and senior-leadership levels.
2. Identify the organisation's serious risks, critical controls and legal reporting requirements by jurisdiction.
3. Choose a small balanced set of leading and lagging indicators linked to those decisions.
4. Write a definition for each measure, including source, denominator, owner, frequency and escalation rule.
5. Check the quality and completeness of site, people, risk, action and incident data before comparing locations.
6. Build the path from exception to underlying record and evidence.
7. Test the view with site managers, safety representatives and senior users.
8. Review whether the dashboard changes decisions and closes risk, then remove measures that do not.

Primary sources

• Health and Safety Authority: Safety and Health Management Systems
• Health and Safety Authority: Safety Management Made Simple
• Health and Safety Authority: Safety Statement and Risk Assessment
• HSE: Check — monitor and report health and safety performance
• HSE: Leadership checklist
• OSHA: Recommended Practices for Safety and Health Programs
• OSHA: Management Leadership
• OSHA: Hazard Identification and Assessment

This article provides general operational guidance for organisations in Ireland, Great Britain and the United States. Apply the legislation, regulator requirements and competent advice relevant to each workplace and jurisdiction.