By Secret Customer Australia
Mystery shopping records what happens during a planned customer interaction against defined criteria. Customer surveys collect customers’ reported experiences and evaluations. Customer satisfaction measures focus on satisfaction; Net Promoter Score focuses on likelihood to recommend. These methods answer different questions. Used together, they can help businesses understand both service delivery and how customers interpret it.
The choice starts with the decision you need to make. If you want to know whether a required explanation was given, ask for evidence of the interaction. If you want to know whether customers felt confident or valued, their own responses matter.
Compare the measures
| Method | Core question | Source | Useful for | Main limitation |
|---|---|---|---|---|
| Mystery shopping | Were the defined behaviours and process steps delivered? | A trained assessor completing a planned scenario. | Specific service standards, enquiry handling, explanation, follow-up and scenario outcomes. | A sample of planned encounters cannot represent every customer or establish the cause of a wider business result. |
| Customer satisfaction or CSAT | How satisfied was the responding customer? | Customers reporting their evaluation of an interaction or relationship. | Understanding perceived performance and gathering explanations in customers’ own words. | Responses depend on who answers, expectations, question design and timing. |
| Net Promoter Score or NPS | How likely is the responding customer to recommend the business? | Customers answering a recommendation question. | Tracking advocacy signals and identifying issues for follow-up. | The headline score does not explain which service behaviours occurred or what caused the rating. |
| Customer effort measure | How easy or difficult did the customer find the task? | Customers responding to a consistently defined effort or ease question. | Finding friction, repeated information, difficult handovers and resolution barriers. | Scales and scoring conventions vary; a score needs context and consistent interpretation. |
A survey is a collection method. CSAT, NPS and customer effort are different measures that may appear in a survey; they should not be treated as interchangeable scores.
What mystery shopping can reveal
A well-designed mystery shop can record whether a greeting occurred, which needs questions were asked, how a product was explained, whether a suitable next step was offered and whether a promised follow-up arrived.
These observations can support targeted coaching. “The employee did not explain the cancellation conditions” identifies a more specific issue than an overall satisfaction rating.
Some observations still require judgement. Defining behavioural anchors and checking narrative evidence helps make those judgements more consistent. Read our complete guide to mystery shopping for the wider context.
What customer surveys add
A customer can report whether the explanation felt clear, whether the experience was fair and whether the outcome met their needs. These perceptions are part of the experience, even when all required process steps occurred.
Open comments can reveal an issue the assessment questionnaire did not anticipate. However, survey respondents may differ from non-respondents, and recall can vary. Record who was invited, when responses were collected and the response volume before drawing broad conclusions.
How NPS differs from a satisfaction score
NPS uses a 0–10 recommendation scale. Responses of 9 or 10 are classified as promoters, 7 or 8 as passives and 0–6 as detractors. The score is the percentage of promoters minus the percentage of detractors; it can range from −100 to +100.
Illustrative example: if 50% of respondents are promoters and 20% are detractors, NPS is +30. It is not 30% satisfaction. The remaining respondents still form part of the denominator.
Bain & Company explains the NPS calculation. The score describes recommendation responses; follow-up questions and other evidence are needed to investigate the reasons.
How CSAT should be interpreted
CSAT asks about satisfaction with a defined experience, product or relationship. A business may calculate the percentage selecting specified positive categories, but it must state the question, scale, categories counted and denominator.
For example, counting “satisfied” and “very satisfied” on a five-category scale is a different convention from reporting an average rating. Avoid comparing them as though they are the same measure.
Why the results can disagree
| Pattern | Possible explanation | Next check |
|---|---|---|
| Strong assessment results but weak satisfaction. | Required steps occurred, but the process remained slow, confusing or unsuitable. | Read customer comments and test the end-to-end journey. |
| Weak assessment results but strong satisfaction. | The customer liked the outcome even though an important process step was missed. | Investigate the missing behaviour rather than using satisfaction as proof it occurred. |
| Good local service but weak recommendation. | Price, product, brand history or another part of the relationship influenced the rating. | Separate interaction-level observations from relationship-level feedback. |
| Fast responses but high reported effort. | The first reply was quick, while resolution required repeated contact or handovers. | Measure completion and total effort alongside initial response time. |
These are diagnostic possibilities, not established causes. Differences may also reflect timing, sampling or the customers and scenarios included.
Choose the method for the management question
- Checking whether training is applied: use observable assessment criteria, then examine customer and operational outcomes.
- Understanding dissatisfaction: collect customer feedback and investigate relevant journeys.
- Monitoring recommendation: use a consistent NPS approach with follow-up questions and sample context.
- Reducing avoidable effort: combine customers’ reported effort with journey tests, repeat-contact records and completion data.
Combine evidence without combining unlike scores
A useful dashboard can show mystery shopping results, customer feedback, NPS and operational measures beside each other. Align the reporting period, journey definitions and location labels so managers can investigate patterns.
Do not average a mystery shopping percentage with NPS to create an unexplained “overall experience score”. The scales, constructs and samples differ. Keep each measure’s meaning visible.
A practical cycle is to identify a concern in customer feedback, test relevant service behaviours, implement an appropriate change and check both delivery and customer outcomes afterwards. See how mystery shopping contributes to business intelligence.
Frequently asked questions
Can mystery shopping replace customer surveys?
It can answer some operational questions that surveys do not answer reliably, but it does not replace customers’ own accounts of satisfaction, effort or confidence.
Can NPS tell us which employees need coaching?
The score alone cannot establish which behaviour caused a response. Specific interaction evidence and appropriate context are needed before making individual conclusions.
Is one method always more accurate?
Accuracy depends on the question, design, sample and execution. A carefully delivered measure can still be unsuitable for the decision you want to make.
Related guides
Choose a measurement approach for your business
Secret Customer offers mystery shopping services, customer survey services and NPS services. Talk to our team about the questions your measurement program needs to answer.
Sources and scope
This article adapts Secret Customer’s comparative methodology references on mystery shopping, customer surveys, NPS and behavioural measurement. The NPS definition follows Bain & Company’s published calculation. Examples are illustrative, and the article does not report a Secret Customer national benchmark.
