A favourable dashboard can still hide a poor result
A distributor reports excellent delivery performance because its system records the date goods leave the warehouse. Customers report late arrival, incomplete orders and missing certificates. The dashboard is measuring a real event, but it does not fully answer whether the customer received the agreed service.
Performance evaluation asks the organisation to examine what it knows about its quality system and what conclusions that information supports. This is the topic associated with clause 9 in the requirements overview. The examples below explain useful practice; precise ISO 9001:2026 criteria need the official standard.
Start with a management question
Before choosing a measure, state the decision it should support. Management may need to know whether orders are complete on arrival, whether a new picking method is effective or whether supplier errors are reaching customers. These questions require different evidence.
Define each measure’s boundaries. An on time result depends on the agreed date, the event being counted and the handling of changes. If the customer requests a later date, preserve enough information to distinguish a genuine change from an internal rescheduling that makes performance appear better.
Use a meaningful denominator where appropriate. Five incorrect orders have a different implication among fifty shipments than among fifty thousand. Also consider severity. A small number of wrong material certificates could matter more than many minor packaging defects, depending on the products and customer requirements.
Check the information before drawing conclusions
Follow a few dashboard entries back to their source records. Establish whether warehouse staff, the carrier and customer service use the same order identifiers. Missing links can produce duplicate complaints or wrongly associate a failure with an unrelated shipment.
Look for changes in collection methods. A reduction in complaints following a website redesign might reflect a less accessible reporting channel. A rise in recorded picking errors after introducing scanning might reflect improved detection rather than deteriorating performance.
These possibilities do not invalidate the information. They identify questions that should be resolved before management acts. Record important limitations alongside the conclusion, especially where a small sample or incomplete response could materially affect the decision.
Use customer feedback as more than a score
A satisfaction survey offers one view of customer perception. Other evidence includes complaints, repeat enquiries, returns, account discussions and the reasons customers stop ordering. A distributor’s largest customer may provide detailed performance reviews that reveal issues absent from a short general survey.
Consider who responded and who did not. Results from customers with easy orders may not describe customers using a complex delivery service. Breaking information into relevant segments can show where an overall average conceals a recurring problem.
The APG Customer Feedback paper offers historical auditing guidance on this subject. Its publication belongs to the 2015 collection and does not establish the current edition’s exact requirements. Here it supports asking how feedback becomes an informed assessment rather than merely a collected statistic.
Let internal audits test the process
An internal audit can follow an order from the customer’s requirement through purchasing, receipt, picking and delivery. Compare evidence with the applicable criteria, including the organisation’s own commitments and the authorised standard where relevant.
Choose samples that reflect the audit purpose. A changed order, a shipment involving an external warehouse and a complaint may reveal different interfaces. Sampling only straightforward orders can provide an incomplete picture, while sampling only failures can also distort the conclusion if presented as typical performance.
Preserve the scope and limits of what was examined. An auditor can report what the evidence supports without claiming that a small sample proves every shipment was correct. Findings should help management understand both conformity and the process conditions behind the observed result.
Bring the evidence into system decisions
Individual teams may resolve local picking mistakes. Management review is the place to consider wider questions such as whether the delivery promise, carrier arrangement or staffing model remains suitable. The management review article develops how to turn evidence into decisions and follow through.
Avoid presenting disconnected slides with no implications. If supplier defects are increasing, explain the affected orders, customer consequences, existing response and decision needed. Management can then compare choices, including supplier development, alternative sourcing or revised verification.
A decision to continue the current approach can be legitimate when supported by the evidence. The record should make the reasoning understandable, particularly where uncertainty or resource constraints are significant.
Evaluate the effects of decisions
Suppose the distributor adds a check for required certificates before dispatch. Subsequent evaluation should examine whether customers now receive usable certificates with the correct goods. Counting the number of completed checks would show activity but leave the customer outcome unresolved.
Also watch for side effects. A manual check may create a dispatch bottleneck, or staff may tick the field without checking the attachment. Follow a sample into actual delivery and ask employees how the control behaves during busy periods.
Performance evaluation is most useful as a repeating connection between questions, trustworthy information, proportionate conclusions and action. The chosen charts and reports should make that connection clearer. They should help people decide what to preserve, what to investigate and what to change, with the limits of the available evidence visible.