How root-cause analysis separates symptoms, unusual circumstances, controllability, and corrective action.
Root-cause analysis prevents management from fixing the wrong problem. A variance, missed target, or unusual metric is only a signal. The case response should separate symptom from cause, identify whether the factor is controllable, and recommend the next analysis or action that would improve performance.
Root-cause analysis belongs in Management Accounting and Performance when a variance, KPI movement, benchmark gap, or unusual result needs explanation before management can choose a fair and effective response.
| Coverage area | Performance Management question |
|---|---|
| Symptom versus cause | What underlying process, capacity, market, data, or behaviour driver explains the result? |
| Unusual circumstance | Is the cause temporary, external, non-recurring, or evidence of operating weakness? |
| Controllability | Which factors should affect manager evaluation, targets, or remedial action? |
| Evidence need | What data, comparison, trend, or operational evidence would confirm the cause? |
| Recommendation | What next analysis or action can prove the cause and improve performance? |
The same symptom can have different causes.
| Symptom | Possible root causes |
|---|---|
| Revenue below target | Lower demand, price pressure, poor sales execution, lost key customer, channel issue, unrealistic target. |
| Gross margin decline | Discounting, input cost increase, product mix shift, waste, rework, supplier issue, pricing error. |
| Labour variance unfavourable | Training gap, scheduling issue, overtime, bottleneck, rework, absenteeism, unrealistic standard. |
| Customer complaints rise | Quality defect, delayed fulfilment, service capacity, unclear communication, product mismatch. |
| Cost per unit increases | Lower volume, fixed-cost absorption, inefficient process, higher input price, maintenance downtime. |
| Dashboard metrics conflict | Measure definition problem, aggregation issue, trade-off between cost and quality, timing difference. |
Unusual circumstances can explain a result without eliminating the need for action.
| Circumstance | Performance evaluation effect | Management action |
|---|---|---|
| One-time supplier disruption | May justify target adjustment or separate reporting. | Develop supplier contingency or inventory policy. |
| Start-up or implementation period | Early inefficiency may be expected. | Use transition targets and monitor learning curve. |
| Market shock or regulatory change | Some variance may be uncontrollable. | Revise forecast, risk assessment, and strategy. |
| Key staff turnover | Short-term service or quality decline may be explainable. | Address succession, training, and workload. |
| System conversion | Data or process results may be temporarily unreliable. | Reconcile data and add temporary controls. |
| Weather, strike, or physical disruption | Operations may be constrained. | Evaluate insurance, continuity, and customer communication. |
The response should state whether the circumstance is temporary or recurring. A recurring “unusual” issue is no longer unusual; it should be built into planning and risk management.
Root-cause conclusions should distinguish controllable and uncontrollable factors.
| Factor | Usually controllable? | Response implication |
|---|---|---|
| Staff scheduling | Often yes | Investigate staffing model, overtime approval, and demand forecasting. |
| Market-wide input price increase | Often no at manager level | Adjust standard or evaluate supplier strategy rather than blame operations manager. |
| Product defect rate | Often yes or partly yes | Trace to design, supplier, process control, or training. |
| Customer mix shift | Partly | Evaluate sales strategy, pricing, and customer profitability. |
| Exchange-rate movement | Often no at operations level | Consider hedging, pricing policy, or separate reporting. |
| Outdated standard | Management controllable | Update planning assumptions and evaluation target. |
When the case does not prove the cause, ask for evidence that would resolve the uncertainty.
| Suspected cause | Evidence that would confirm it |
|---|---|
| Price discounting drove sales | Average selling price by product, customer, and channel. |
| Product mix hurt margin | Contribution margin by product and sales-mix trend. |
| Process inefficiency caused costs | Cycle time, rework, downtime, scrap, and bottleneck data. |
| Supplier quality caused defects | Supplier defect rate, inspection results, delivery delays, and returns. |
| Target was unrealistic | Historical trend, benchmark, capacity, staffing, and market data. |
| Incentive caused behaviour | Bonus formula, manager actions, timing of transactions, and measure manipulation signs. |
Use this sequence: identify the symptom, list plausible causes, use case facts to rank causes, separate controllable from uncontrollable factors, identify unusual circumstances, recommend confirming analysis, and state the remedial action if the leading cause is confirmed.
Avoid overclaiming. If the facts support only a likely cause, say so and recommend the specific analysis needed.
| Pitfall | Correction |
|---|---|
| Treating the variance as the cause. | Explain what created the variance. |
| Assuming an unfavourable result means poor management. | Test controllability, target realism, and unusual circumstances. |
| Recommending action before diagnosis. | Identify the evidence needed to confirm the cause. |
| Ignoring conflicting metrics. | Explain trade-offs and aggregation issues. |
| Treating a recurring issue as unusual. | Build recurring conditions into standards, forecasts, and risk management. |