Opens in a new tab
How mature is your risk management?
Take the Free AssessmentTake Assessment

Risk Management

Risk Matrix: A 5×5 Template, Examples and How to Use One Well

Keith Ricketts-Clew-GRC

Keith Ricketts

November 11, 2026
Time to read: 10 minutes

Summarize this article with:

ClaudeChatGPTGoogle GeminiPerplexityGrok
Under 10%
of random hazard pairs a typical risk matrix can correctly compare, per Cox
4
documented limitations every matrix user should know
25
cells in a 5x5 risk matrix, each needing a deliberate rating
Didn't have time to read - TL;DR
3 key takeaways
1

A risk matrix is a shared language, not a calculator. It turns judgements about likelihood and consequence into a rating everyone reads the same way, so the definitions matter more than the maths.

2

Rate the cell, don’t just multiply. Scores from 1 to 25 are a convenience; assign each cell’s rating deliberately and calibrate it to your risk appetite.

3

Use one matrix everywhere. Matrices that differ between sites or projects produce ratings nobody can compare or add up.

The board pack has a risk summary on page four. Supplier failure appears twice: rated high by the infrastructure division and medium by the construction division. It is the same supplier, the same contract and the same exposure. The divisions simply use different versions of the risk matrix, with different definitions of “likely” and different ideas of what a major consequence costs.

The summary reports one high and one medium. Nobody on the board can tell which is right, because neither rating means the same thing as the other.

The matrix is the most widely used tool in risk management, and one of the most quietly inconsistent. This guide covers what it is, how a 5×5 matrix works, a worked example, a template you can copy, how to calibrate the matrix to your risk appetite, how to use one across an organisation, and the limits every user should know.

What Is a Risk Matrix?

A risk matrix is a grid that rates risks by combining how likely they are with how severe their consequences would be. One axis carries a likelihood scale and the other a consequence scale, and each cell where they meet carries a rating, usually low, medium, high or extreme, that tells people how urgently the risk needs attention and who must act on it.

Matrices spread partly through risk management standards. Cox’s 2008 paper in Risk Analysis names AS/NZS 4360, the forerunner of today’s Australian standard, among the standards that drove their adoption.

How a 5×5 Risk Matrix Works

A 5×5 matrix has five levels of likelihood and five levels of consequence, which gives 25 cells. The likelihood scale needs a guide people can test against, ideally a frequency. Without one, “likely” means whatever the person filling in the register thinks it means.

LevelLikelihoodGuide
1RareCould occur only in exceptional circumstances, less than once in 20 years
2UnlikelyCould occur at some point, once in 5 to 20 years
3PossibleMight occur, once in 1 to 5 years
4LikelyWill probably occur, about once a year
5Almost certainExpected to occur, several times a year

The consequence scale needs the same discipline, and it needs to cover more than safety. That is set out in the example below. Where the two scales meet, each cell gets a rating.

Illustrative bands from this guide. Calibrate them to your own risk appetite.

The cell ratings are where most matrices go wrong. Many guides multiply likelihood by consequence and band the score: 1 to 4 low, 5 to 9 medium, 10 to 16 high, 17 to 25 extreme. It is a useful shortcut, but it treats judgements as if they were measurements. A rare but severe risk scores 5, the same as an almost certain but insignificant one, and few boards would treat those two the same.

Assign each cell’s rating deliberately instead, starting with the most severe consequences, then check that the grid reads sensibly. In the matrix above, a score of 8 is high where the consequence is major and medium where it is minor, because the organisation cares more about severity than frequency.

Rule of thumb

Rate the cell, not the sum. The score is a convenience; the rating is a decision.

Risk Matrix Example

Here is one risk rated end to end, using a consequence scale that covers five categories. The financial bands suit a mid-sized organisation; set yours to your own size and appetite.

LevelSafetyFinancialReputationEnvironmentCompliance
1 InsignificantFirst aid onlyUnder $50,000A local complaint, no mediaContained on site, no lasting effectMinor administrative breach
2 MinorMedical treatment$50,000 to $500,000Local media coverageMinor off-site impact, quickly cleaned upBreach reported to a regulator, no action taken
3 ModerateLost-time injury$500,000 to $2 millionState or national media for a dayOff-site harm needing remediationRegulator investigation or formal warning
4 MajorSerious injury or permanent disability$2 million to $10 millionSustained national coverageSignificant harm with lasting effectsProsecution or enforceable undertaking
5 SevereFatality or multiple serious injuriesOver $10 millionLong-term loss of trustMajor, long-term environmental damageLoss of licence or major prosecution

The risk is a key supplier failing midway through a major infrastructure contract. The supplier has missed two payments to its own subcontractors this quarter, so likelihood is rated possible (3). Across the consequence categories, safety is minor, environment insignificant, compliance minor and reputation moderate, but the financial consequence of delay is about $3 million, which is major (4).

Rate on the most severe credible consequence, not an average. Possible and major meet in a high cell. Had both divisions in the opening example used this matrix, they would both have reached the same answer.

Risk Matrix Template

To use this risk matrix template, copy the likelihood table, the consequence table and the grid below, then replace the illustrative bands with your own.

LikelihoodInsignificantMinorModerateMajorSevere
Almost certain (5)MediumMediumHighExtremeExtreme
Likely (4)LowMediumHighExtremeExtreme
Possible (3)LowMediumMediumHighExtreme
Unlikely (2)LowLowMediumHighHigh
Rare (1)LowLowLowMediumHigh

Then define what each rating means in practice.

  • Low risks are managed by routine procedures and reviewed annually.
  • Medium risks have a named owner and are reviewed each quarter.
  • High risks need senior management attention and a treatment plan with dates.
  • Extreme risks go to the executive straight away and are reported to the board.

3×3, 4×4 or 5×5: Which Size to Use

A 3×3 matrix is quick and coarse. It suits a single site’s daily hazard checks, where speed matters more than nuance. A 4×4 matrix removes the middle column, which forces assessors to choose a side instead of defaulting to moderate. A 5×5 matrix is the most common choice for an organisation-wide register, because it separates rare catastrophes from frequent nuisances without pretending to more precision than the inputs can support.

More cells give more apparent precision, not more accuracy. The inputs are still judgements, so choose the smallest matrix that separates the risks your board needs to see apart.

Risk Matrix vs Risk Heat Map

The two terms are often used interchangeably, but they do different jobs. The risk matrix is the rating tool: the scales and the grid you use to rate each risk. A risk heat map is a report: the same grid with your actual risks plotted on it, so a board can see at a glance where they cluster.

Build the matrix first. A heat map is only as meaningful as the matrix underneath it, and a heat map drawn from several different matrices is the board pack problem in picture form. A useful heat map also shows movement, such as where each top risk sat last quarter, and names the risks rather than just counting them. For what else belongs on that page, see what directors actually need to see in a risk report.

How to Calibrate a Risk Matrix to Your Risk Appetite

A risk matrix is only as good as its fit with what the organisation is willing to accept. Comcover’s guidance on risk appetite makes the point directly: if likelihood and consequence tables are not calibrated, the ratings they produce can drive responses that are too light or too heavy.

Three steps calibrate a matrix.

  • Set the consequence bands from your tolerances. If the risk appetite statement says forecast variance above 5% goes to the audit and risk committee, the financial band where that happens should sit at major or above.
  • Weight the grid towards what you will not accept. If the organisation has no appetite for fatalities, a severe safety consequence should rate at least high however rare it is, which is why the bottom-right cell above is high.
  • Test it against real events. Rate three or four incidents from the last few years, and check that the ratings match how seriously the organisation actually treated them.

The appetite side of this is covered in writing a risk appetite statement.

Using One Matrix Across the Organisation

Back to the board pack. Two divisions rated the same supplier differently because each had built its own matrix. The fix is organisational, not mathematical.

  • Set one set of scales centrally, with consequence bands sized to the organisation rather than to each project.
  • Let local teams add guidance and examples, but not change the definitions.
  • Run calibration sessions where divisions rate the same scenario, and discuss any differences.
  • Report every rating on the same scale, so the board can compare and add them up.

This is hard to enforce in spreadsheets, where every copy of the register can carry its own version of the matrix. Purpose-built risk management software applies one matrix across every register, with dashboards and board-level reports drawn from the same ratings.

Two divisions with two matrices are not reporting two views of risk. They are reporting in two languages.

The Limits of Risk Matrices, and How to Work Around Them

Cox’s 2008 analysis set out four limitations of typical risk matrices. Each has a practical workaround.

  • Poor resolution. Typical matrices can correctly compare fewer than 10% of randomly selected pairs of hazards, and they can give the same rating to very different risks. Use the matrix to triage, then look at the risks inside each band rather than ranking them by score.
  • Ranking errors. A matrix can rate a smaller risk higher than a larger one. Review your most severe consequences individually, whatever cell they land in.
  • Poor guide to resource allocation. A colour cannot tell you where the next dollar of treatment does the most good. Decide treatment on the cost and effect of controls, not on the rating alone.
  • Ambiguous inputs and outputs. Different people can rate the same risk differently. Written guides for every level and regular calibration sessions narrow the gap.

None of this means abandoning the matrix. Cox’s own conclusion is to use matrices with caution, and with clear explanations of the judgements built into them.

The matrix rates a risk at one moment. Two things keep that rating honest between reviews: evidence that the controls work, which moves the residual rating, covered in how to assess control effectiveness, and indicators that show when a rating is drifting, covered in key risk indicators. Together they form part of the traceable link from an organisation’s objectives through its risks, controls, actions and assurance to the indicators that show whether it is performing, and the matrix is where every risk in that chain gets its rating.

Is There an Australian Standard Risk Matrix?

No. The standard people usually mean is AS ISO 31000:2018, Risk management: Guidelines, which is listed as current by Intertek Inform, the publisher of Australian standards under licence. It is identical to the international ISO 31000:2018, and it descends from AS/NZS 4360, first published in 1995, by way of AS/NZS ISO 31000:2009.

The standard sets out principles, a framework and a process for managing risk. It does not prescribe a particular matrix, so organisations design their own to fit their risks and appetite, which is exactly why matrices differ so much between organisations.

Frequently Asked Questions

A risk matrix is a grid that rates a risk by combining its likelihood with the severity of its consequences. Each cell carries a rating, such as low, medium, high or extreme, that tells people how urgently the risk needs attention and who must act on it.

A 5×5 risk matrix has five levels of likelihood, from rare to almost certain, and five levels of consequence, from insignificant to severe. The result is 25 cells, each rated deliberately. It is the most common size for an organisation-wide risk register.

Define the likelihood and consequence scales, rate a risk’s likelihood and its most severe credible consequence, and read the rating from the cell where they meet. Then act on the response agreed for that rating, and review the rating when circumstances change.

Research by Cox (2008) identified four: poor resolution, ranking errors, a poor guide to where to spend on treatment, and ambiguous inputs and outputs. Matrices are still useful for triage, provided the scales are clearly defined and the judgements behind each rating are recorded.

No. AS ISO 31000:2018, the current Australian risk management standard, sets out principles and a process but does not prescribe a particular matrix. Organisations design their own to fit their risks and appetite.

Put your matrix to work. Clew is a risk and assurance platform built around the Golden Thread, so a risk’s rating connects to the objective it threatens, the controls that manage it and the indicators that show whether it is moving. Risk and assurance teams in government, healthcare, construction, infrastructure and utilities, mining and resources, and transport use Clew to keep their risks, controls and indicators connected to the objectives they protect. Find out how mature your approach is with the free risk maturity assessment, or explore Clew’s risk management software.

Related Resources

Keith Ricketts-Clew Risk Software

Meet the author

Keith Ricketts

VP Of Marketing

25 years with global software vendors in Cyber Security, Risk and ERP. Blends strategic marketing, agility and a proven record of results.
Connect on LinkedIn

In this article

How mature is your risk management?

Take the free assessment and get a personalised report in minutes.
Take the assessment