Solving today’s problem and preventing recurrence
An equipment rental company sends the wrong accessory with a machine. It arranges a replacement delivery and the customer can continue working. That response addresses the immediate problem. It does not yet explain how the wrong accessory was selected or whether another order is exposed to the same failure.
Improvement is the topic of clause 10 in the requirements guide. This article explains a practical learning process without reproducing the standard’s requirements. Obtain the official ISO 9001:2026 text for exact duties, terminology and audit criteria.
Establish what happened and what is affected
Before discussing causes, identify the requirement and the evidence of the failure. The customer requested a specified accessory, and dispatch supplied another item. Confirm the order, the item identifiers and the delivery evidence. Avoid beginning with a conclusion such as careless employee before the facts are understood.
Consider the extent. Other orders may have been picked from the same storage location or prepared using the same catalogue information. The company may need to check those orders as part of protecting customers. The appropriate response depends on the actual consequence and reach of the problem.
Preserve records that might change during investigation. If a stock description is corrected immediately, retain enough information to understand what the picker saw at the time. This supports reliable analysis while allowing the business to contain the issue promptly.
Look for a cause that explains the evidence
Interview the people involved and observe the relevant work. In this example, two similar accessories may share almost identical descriptions, storage positions and images. A picker could follow the normal method and still choose the wrong item because the distinguishing information is unclear.
Another possibility is that the sales system substituted a default accessory when the customer’s requested item was unavailable. That would require a different response. Investigation should discriminate between plausible explanations rather than select the first familiar label.
Tools such as repeated why questions or a cause diagram can structure thinking. They do not guarantee a correct conclusion. Test the proposed cause against the timing, affected orders and unaffected cases. If it cannot explain the observed pattern, gather more evidence before designing the long term action.
Select action that addresses the mechanism
If ambiguity in item identification is the cause, improve the distinguishing information and test it with users. Moving items apart might help, but the same ambiguity could remain in the booking and dispatch systems. Consider the whole path from order creation to return processing.
If a system substitution caused the failure, changing the picking label alone will leave the original decision uncontrolled. The company may need explicit approval before replacing the requested accessory, with a dependable route for communicating the agreement to dispatch.
Training can support either change, particularly when people need a new skill or decision rule. It should not become the default response when the process design makes the wrong action easy and the correct action difficult to recognise.
Distinguish implementation from effectiveness
An action owner may provide a revised catalogue, updated labels and briefing records. These show that planned changes were implemented. The next question is whether the changes prevent the relevant failure under actual working conditions.
Choose an effectiveness check linked to the mechanism. Review later orders containing the similar accessories, observe how pickers identify them and examine any substitutions. Include busy periods or less experienced staff if those conditions are relevant to the original exposure.
A period with no complaints provides some information, but its meaning depends on the number and type of opportunities for the failure to recur. No recurrence across two simple orders gives less assurance than consistent correct selection across varied representative orders. Avoid imposing an arbitrary universal monitoring period as though ISO specifies it.
Improve without waiting for a failure
Employees may notice that accessory compatibility checks require repeated calls to a technical specialist. Even where the current result is correct, a clearer compatibility guide could reduce delay and dependence on one person. This is a potential improvement opportunity rather than necessarily a nonconformity.
Evaluate the expected benefit and possible consequences. A simplified guide could become misleading if it hides important exceptions. Pilot it on representative equipment and obtain technical review before making it the normal decision aid.
The risks and opportunities guide connects this kind of choice to planning. Improvement can strengthen capability and service, as well as respond to a specific failure.
Learn at system level
A recurring allocation mistake might reveal more than a weak picking process. Management may find that product information changes are approved without considering their effect on sales, dispatch and returns. Addressing that interface can prevent several different failure types.
The APG Improvement paper provides historical audit guidance for the 2015 edition and considers improvement in relation to system performance. Its advisory content should not be treated as the 2026 clause text or as a prescribed improvement quota.
At the end of an improvement activity, the company should be able to explain the original issue, the supporting evidence, the chosen response and the effect observed. Where the effect remains uncertain, say so and define the next review. Honest uncertainty is more useful than a closed action that leaves the failure mechanism unexplained.