A risk and control matrix (RCM) for accounts payable lists, row by row, each thing that could go wrong in paying suppliers, how serious it is, and the control that should stop or catch it. A typical row reads: risk “unauthorised payments”; impact and likelihood rated 1 to 5 and multiplied (inherent risk); control objective “prevent unauthorised transactions”; control “an authority matrix with value-based limits for approving payments”; then the control’s method, type, frequency, owner and the test you will run. You build it from the process walkthrough before testing starts, and it becomes the document you work from for the rest of the audit.
Key points
- An RCM puts every identified risk in one column and the control that should mitigate it in the next, linked to the business objective.
- Build it after the process walkthrough and before testing. Agree it with management, the board or the audit committee before you rely on it.
- Rate each risk for impact and likelihood (1 to 5 each); the product is the inherent risk, and its position on the heat map gives a severity of high, medium or low.
- Write the control description as a positive statement of what should happen, such as “Authority matrix is defined along with value-based limits for approving payments.”
- Record each control’s attributes: COSO objective and component, manual or automated, directive, preventive or detective, key or not, assertion, owner, fraud risk, frequency, the evidence (information produced by the entity) and the test procedure.
- During the audit the same row records population, sample, deviations, control effectiveness, residual risk and the control status.
What a risk and control matrix is, and why accounts payable is a good place to start
A risk and control matrix is a table that an auditor (and risk professionals) use to assess risks and to evaluate whether the controls put in place to reduce them are working. One side of the table lists the risks. The other side lists the controls expected to deal with each risk and to help the department meet its objective.
In the course we call the RCM the auditor’s “Bible” for each audit. It matters for two reasons. First, it stops you from going straight from the walkthrough into testing. If you do that, you may miss important controls, because the RCM covers both what you saw in the process and the ideal controls that any organisation should have. Second, it is where your results go. Whether a control is effective, ineffective or badly designed, you record it in the RCM first, and the audit report summarises it from there.
Accounts payable is a good process to learn on. The risks are easy to picture (paying the wrong party, paying without authority, paying too early or too late) and the standard controls are much the same across organisations. That makes it easier to see the gaps that are particular to the company you are auditing.
The columns of an accounts payable RCM
The course works through this layout, column by column:
| Column | What goes in it | Accounts payable example |
|---|---|---|
| Process | The business unit or primary process | Accounts payable |
| Sub-process | The part of the process the row covers | Policies relating to accounts payable |
| Risk description | What could go wrong | Unauthorised transactions |
| Impact and likelihood | Each rated on a scale of 1 to 5, from your risk assessment | Impact 5 × likelihood 3 = 15 |
| Inherent risk and severity | The risk before any control, and where it sits on the heat map | 15, high |
| Control objective | The risk turned around into an objective | Prevent unauthorised transactions |
| Control description | A positive statement of the control that gives assurance the objective is met | Authority matrix is defined along with value-based limits for approving payments |
| COSO objective and component | Which COSO objective the control serves and which component it belongs to | Safeguarding of assets; control environment |
| Control method | Manual or automated | Manual (approvals are signed on paper vouchers) |
| Control type | Directive, preventive, detective or corrective | Directive (a direction from the board on who may approve, and up to what limit) |
| Criticality | Key control or not | Key: a failure leads to unauthorised payments |
| Control assertion | Completeness, existence, cut-off, classification or accuracy | Set by what the control protects |
| Responsibility | The person who performs the control; you will interview them | Named control owner |
| Fraud risk | Yes or no | Yes: unauthorised transactions carry fraud risk |
| Frequency | How often the control runs; this drives your sample | Ongoing, daily, monthly, quarterly |
| Information produced by the entity | The document or report you will examine | Authority matrix with specimen signatures and value limits |
| Test procedure | What you will do to evaluate the control | Verification and analysis of that document or report |
| Control weight | 1 to 5, by how critical the risk is | 5 |
How to prepare an accounts payable RCM, step by step
- Understand the process first. Do the process walkthrough of the accounts payable team. While you do it, keep the RCM template in mind so you collect everything you will need later: objectives, activities, who does what, and which documents and reports exist.
- Document the process. Prepare a process flowchart or some other process document, from invoice receipt through to payment.
- Identify the risks. At each step, ask what could go wrong. This takes professional judgement, and it improves with experience and with knowing how other companies run the same process. Add the standard controls every organisation should have, then look for what is different in this department compared with a similar one elsewhere.
- Rate each risk. Give impact and likelihood a score of 1 to 5 and multiply them to get the inherent risk. Place it on the heat map to get a severity of high, medium or low.
- Set the control objective. Turn the risk around: for the risk “unauthorised transactions”, the objective is “prevent unauthorised transactions”.
- Describe the control. Write what the company does today as a positive statement. Then check it against the ideal control for that risk, because the control the company runs may not be enough.
- Record the attributes. COSO objective and component, method, type, criticality, assertion, owner, fraud risk, frequency and weight.
- Plan the test. Name the evidence (the information produced by the entity) and the test procedure you will use to evaluate the control.
- Agree the RCM. Discuss it with management, the board of directors or the audit committee and get their confirmation before you start testing.
- Keep it up to date. As fieldwork uncovers steps and controls that you did not see in the walkthrough, add them to the matrix.
Worked example: four accounts payable rows
This is the accounts payable example used in the course, laid out as four RCM rows. It illustrates the method and is not taken from a real client.
| # | Risk | Control description | Method / type | Frequency | Evidence and test |
|---|---|---|---|---|---|
| 1 | Unauthorised transactions: there is no defined rule on who may approve payments, or up to what value (impact 5 × likelihood 3 = inherent risk 15, high) | Authority matrix is defined along with value-based limits for approving payments | Manual / directive; key control; weight 5 | Ongoing | Authority matrix with specimen signatures and value limits; verify it is approved by the board and current |
| 2 | Users can enter or approve payments in the ERP outside their role, so one person can both create and release a payment | User rights in the ERP are segregated and restricted according to each user’s role, so that transactions pass through the approval levels defined in the system | Automated (the ERP enforces the rights) / preventive; fraud risk: yes | Ongoing; the review of rights is quarterly | ERP user access log, downloaded and analysed: are rights in line with roles and with the authority matrix? |
| 3 | Interest cost due to early payments: wrong payment terms in the system bring the due date forward | Payment terms are correctly captured for each supplier, in line with the agreed purchase order or contract | Manual / preventive | As and when a supplier is set up or terms change | Vendor master report of payment terms; compare a sample of suppliers’ terms with their purchase orders or contracts |
| 4 | Interest cost due to early payments: invoices are paid before the credit period ends | The credit period is fully used before payments are made to suppliers; payment runs pick invoices by due date | Automated / preventive | Each payment run | Payment run reports and invoice due dates; for a sample of payments, compare the payment date with the due date and ask why any were paid early |
Why row 1 is directive and row 2 is preventive. The authority matrix is a policy: the board’s direction on who may approve, and up to what value. That makes it directive. Once that direction is built into the system and every transaction has to pass through the approval levels set there, the control stops an unauthorised approval before it happens. That makes it preventive.
Why row 1 is manual here. In this example the authority matrix is applied offline. Each voucher moves physically from the accounts manager to the CFO or finance controller for signature. If the same approvals were routed through the ERP to different users, the method would be automated.
A common case in practice. The authority matrix is approved and up to date on paper, but the ERP has not been set up to match it. Approval levels in the system may still follow an older version of the matrix, or a finance user may hold rights both to create a vendor and to release a payment. Row 1 looks fine when you read the policy. Row 2 fails, because the system does not enforce what the policy says. This is why the RCM keeps the directive control and the system control as separate rows. It is also why the evidence for row 2 is the user access log, not the policy document.
What a full accounts payable RCM goes on to cover
The four rows above teach the method. A complete matrix for the process usually covers these objectives, each with its own risks and controls:
- Vendor master data: vendors are created and changed only with authorisation. One example control is a call-back to a known contact, taken from existing records, before any change to a vendor’s bank details (preventive, manual).
- Invoice recording: invoices are recorded promptly and accurately, and only for goods and services that were ordered and received. One example control is a system three-way match of invoice, purchase order and goods receipt (preventive, automated).
- Tax: TDS is deducted at the right rate and GST input credit is claimed correctly. One example control is a monthly reconciliation of input credit in the books to GSTR-2B (detective).
- Timely payment: payments go out on time and available discounts are captured.
- Returns, debit notes and advances: all authorised items are recorded, and only those.
- Complete and accurate payables: one example control is a monthly reconciliation of the AP sub-ledger to the general ledger (detective, manual).
- Payables records and data: they are safeguarded.
Writing risk and control descriptions that work
- State the control as a positive statement of what happens, not as a question or a recommendation. “Cash is disbursed after verification of supporting documents and approval” is a control description. “Check whether approvals are obtained” is a test step.
- Link each control to its risk. The control objective comes from the risk, and the control should be what gives assurance that the objective is met.
- Describe what the company does today, then challenge it. The control the company runs now may not be the control the risk needs.
- Get the frequency right, because it sets your sample. A daily control needs samples spread across the period. A monthly control runs at most 12 times a year.
- Practise. The course’s advice is to practise on the audits you are doing now or, if you are not on an audit, to document the process of any business you know. The quality of your descriptions improves with practice.
After testing: updating the same row
When fieldwork starts, each RCM row gets the results of your testing. The course uses these columns and formulas:
| Field | How it is worked out | Illustration from the course |
|---|---|---|
| Population and sample | Sample % = sample size ÷ population × 100 | 500 transactions tested out of 1,000 |
| Deviation % | Deviations ÷ sample size × 100 (not ÷ population) | 20 deviations in a sample of 100 = 20% |
| Control effectiveness | 1 − deviation % | 1 − 20% = 80% |
| Control test result | Effectiveness × control weight | 80% × 5 = 4 out of 5 |
| Residual risk | The original impact × the likelihood after the control | Impact 4 × likelihood reduced from 5 to 3: 20 falls to 12 |
| Control status | Effective, not effective, design gap, or control missing | With a design gap, look for a compensating control |
A point specific to payables: the formula does not override your judgement. Suppose 98 of 100 payments were approved by the right person, but 2 were approved by someone without authority who could still approve them in the ERP. The control has failed at the system level. The course rates its effectiveness at 0, even though the deviation is only 2%.
Record your conclusion and the reasons for it in the RCM. Link each row to its working papers (by file name, number or location) so that a reviewer can find the evidence.
▶ Watch it explained
Short videos from our internal audit course. Each one covers a single topic, step by step.
📄 Get the template
Payables and Receivables: Risk and Control Matrix (Excel). 27 controls (21 key) for 11 objectives and 16 risks, covering vendor master data, invoice matching, TDS and GST input credit, returns, advances and reconciliation. Each control comes with walkthrough questions, evidence to ask for and test steps.
Procurement: Risk and Control Matrix (Excel). 19 controls (11 key) for 10 objectives and 12 risks for the purchasing steps that come before payables.
✅ Test yourself
Questions from our course quizzes. Click a question to see the answer.
Sarah, an internal auditor, is preparing a risk and control matrix for a manufacturing company. What is the purpose of the RCM for her?
To evaluate the effectiveness of internal controls. The RCM links each risk to the controls meant to reduce it, and rates those controls by how effective they are. It is not mainly a tool for assessing business processes, scoring risks or setting objectives.
Voucher approvals follow the authority matrix, but each level signs a paper voucher. How should the control method be recorded?
Manual, because the approvals happen offline on paper. If the approvals were routed through the ERP to different users, the method would be automated.
A control is reviewed once a month. How does this affect your testing?
A monthly control runs at most 12 times a year, so you need only about 12 samples, one for each occurrence. A daily control needs samples spread across the whole period.
Frequently asked questions
What is an RCM in internal audit?
A risk and control matrix is a table that lists each risk in a process next to the control meant to reduce it, with the risk rating, the control’s attributes and the test the auditor will run. It is prepared after the process walkthrough and used throughout the audit to record how effective each control is.
What columns does an accounts payable RCM have?
Process and sub-process, risk description, impact, likelihood and inherent risk, severity, control objective, control description, COSO objective and component, control method (manual or automated), control type (directive, preventive or detective), criticality, assertion, responsibility, fraud risk, frequency, the evidence (information produced by the entity), the test procedure and a control weight.
What is inherent risk in an RCM?
Inherent risk is the risk before any controls are applied. It is impact multiplied by likelihood, each rated from 1 to 5. Residual risk is the risk after the control: impact stays the same, but an effective control lowers the likelihood.
Who should agree the RCM before testing?
The auditor should discuss the RCM with management, the board of directors or the audit committee and get their confirmation before using it to test controls.
What is the difference between an RCM and an audit checklist?
The RCM maps each risk to its controls and records their ratings and test results. A checklist turns the key controls into questions to work through. Checklists are best built from an RCM, so they test the controls that matter.
Related reading: The internal audit process, step by step · Test of design vs operating effectiveness explained · Preventive vs detective controls: clear examples · Build your own internal audit checklist