Customer Satisfaction: A Guide to Measurement, Analysis, and Improvement

Higher sales do not always mean higher satisfaction. A customer may buy again because there is no better alternative, switching costs are high, or a temporary discount is available—while privately deciding to leave the brand at the first opportunity. Likewise, not every dissatisfied customer complains. Some simply buy less, reduce their use of the service, or quietly switch to a competitor.

For this reason, managers cannot judge customer satisfaction solely by sales, complaint volumes, or a handful of positive social media comments. Customer satisfaction measurement creates managerial value when it is conducted as part of a structured market research process: the right problem is defined, different customer groups are heard, reliable data is collected, and the findings lead to decisions and action.

This guide begins by defining satisfaction and explaining how it differs from customer experience and loyalty. It then covers CSAT, NPS, and CES; quantitative and qualitative methods; questionnaire design; satisfaction driver analysis; and how to turn findings into an improvement plan, with practical examples throughout.

What Is Customer Satisfaction Measurement?

Customer satisfaction measurement is the process of assessing and analyzing how well a customer’s actual experience matches their expectations. Before buying a product or using a service, customers form expectations—for example, that an order will arrive when promised, an app will work without errors, or a support agent will resolve an issue within a reasonable time. After the experience, they compare the brand’s actual performance with those expectations.

Put simply:

Satisfaction results from the customer’s assessment of the gap between “what they expected” and “what they received.”

When performance falls below expectations, dissatisfaction becomes more likely. When it meets expectations, the customer is usually satisfied. When the experience is meaningfully better than expected, satisfaction may be higher. However, this relationship is not mechanical: the importance of each factor, previous experiences, price paid, and usage conditions also shape the customer’s judgment.

There is an important difference between “we feel our customers are satisfied” and measuring customer satisfaction. Frontline employees’ impressions, comments from a few active customers, or a low complaint rate can create an incomplete picture. Rigorous measurement requires a clearly defined target population, an appropriate sample, a standardized instrument, correct timing, and systematic analysis.

For example, an internet service provider may conclude from a low number of complaint calls that customers are satisfied. Yet customer satisfaction research may reveal that, after several outages, some users stop contacting support because they no longer believe follow-up will produce a result. In this case, fewer calls are not a sign of satisfaction; they may indicate declining trust in the problem-resolution process.

Why Is Measuring Customer Satisfaction Important for Businesses?

Measuring satisfaction is not merely about producing a score for a management report. When a study is designed properly, it shows which parts of the experience need improvement, which customer groups are at risk of churn, and where improvement resources should be invested.

Its main applications include:

• Identifying sources of dissatisfaction: It clarifies whether the main issue is price, product quality, waiting time, employee behavior, or process difficulty.
• Reducing customer churn: Combining attitudinal and behavioral data can reveal high-risk groups earlier.
• Increasing customer loyalty: Removing friction points and acting on feedback strengthens the conditions for repeat purchases and brand recommendation.
• Improving products and services: The voice of the customer helps product teams distinguish frequently mentioned requests from genuinely important needs.
• Detecting weaknesses in the customer experience: Customers may be satisfied with the product while payment, delivery, or after-sales service damages the overall experience.
• Prioritizing corrective actions: Not all problems are equally important. Driver analysis shows which improvements are likely to have the greatest impact.
• Strengthening data-driven decision-making: Subjective debate gives way to measurable evidence.
• Monitoring the quality of branches, channels, and contractors: Performance differences across units become visible, although comparisons should control for differences in customer mix.

Management note: A high average satisfaction score is not enough on its own. The response distribution, the share of highly dissatisfied customers, differences between segments, and the reasons behind the score are often more important than the average itself.

For example, when studying different age groups, separate analysis of Generation Z and Generation Alpha can reveal differences in their expectations, behavior, and satisfaction criteria.

[Bar chart: Hypothetical comparison of satisfaction with price, quality, support, and service delivery speed]

Sample chart data—all figures are hypothetical and for educational purposes only:

Factor	Average Satisfaction out of 5
Product quality	4.2
Price	3.7
Support	3.4
Service delivery speed	2.9

This chart is not intended to provide an industry benchmark. It simply illustrates how an acceptable overall score can coexist with a weak touchpoint.

Customer Satisfaction vs. Customer Experience vs. Customer Loyalty

These three concepts are related, but they are not interchangeable.

Concept	Definition	Main Question	Common Metrics	Managerial Use
Customer satisfaction	The customer’s evaluation of an interaction, product, or overall relationship relative to expectations	“How satisfied were you with this experience?”	CSAT, overall satisfaction, satisfaction with specific attributes	Identifying perceived quality and sources of dissatisfaction
Customer experience	The sum of a customer’s perceptions and emotions across all touchpoints throughout the customer journey	“What experience does the customer have while interacting with the brand?”	CES, touchpoint metrics, customer journey analysis, behavioral data	Designing and improving the entire customer journey
Customer loyalty	The customer’s intention and behavior to continue the relationship, buy again, or prefer the brand	“Will the customer stay, purchase again, or recommend the brand?”	Retention rate, repeat purchase, share of wallet, NPS	Managing retention and relationship value

A satisfied customer is not necessarily loyal. Someone may be satisfied with a restaurant but choose a different one each time because they value variety. Conversely, a bank customer may not be highly satisfied but may remain because transferring accounts is difficult or other services depend on the account. To understand the customer’s true relationship with a brand, attitudinal, experiential, and behavioral metrics should therefore be considered together.

[Callout: Satisfied customer vs. loyal customer—a satisfied customer says, “This experience was acceptable”; a loyal customer shows through behavior or preference that, “I will probably choose this brand again.”]

Key Customer Satisfaction Metrics

No single metric can answer every managerial question. Metric selection should begin with the decision the result is intended to support.

Customer Satisfaction Score (CSAT)

CSAT measures a customer’s satisfaction with a particular product, service, or interaction. A typical question is: “Overall, how satisfied were you with the order delivery process?”

If a 1-to-5 scale is used and scores of 4 and 5 are defined as “satisfied”:

CSAT = (Number of 4 and 5 ratings ÷ Total valid responses) × 100

Educational example: Out of 200 valid responses, 150 respondents gave a rating of 4 or 5. CSAT is therefore 75%. This figure is included only to demonstrate the calculation and does not represent any real business.

• Use: Measuring satisfaction after a purchase, support call, branch visit, or use of a feature.
• Advantage: Simple, easy to understand, and suitable for monitoring touchpoints.
• Limitation: Sensitive to the scale used, survey timing, and the definition of a “satisfied” response; it also does not necessarily predict future behavior.

Reports should state the exact question wording, scale, time frame, and calculation rule. Comparing two CSAT results based on different scales or definitions can be misleading.

Net Promoter Score (NPS)

NPS is measured by asking, “How likely are you to recommend this brand/product to a friend or colleague?” on a scale from 0 to 10.

• Scores of 9 and 10: Promoters
• Scores of 7 and 8: Passives
• Scores of 0 to 6: Detractors

This metric is commonly used to assess customers’ relationship attitudes toward a brand and their willingness to recommend it. It is therefore considered one of the metrics used to assess customer loyalty.

NPS = Percentage of Promoters − Percentage of Detractors

Educational example: If 50% of respondents are Promoters, 30% are Passives, and 20% are Detractors, the NPS is +30.

• Use: Monitoring relationship attitudes and willingness to recommend the brand.
• Advantage: A concise metric suitable for tracking over time, provided the measurement method remains consistent.
• Limitation: It does not explain the reason behind the score and is not, by itself, equivalent to behavioral loyalty. Comparisons with unsuitable benchmarks can also distort decisions.

It is useful to follow the NPS question with: “What is the main reason for the score you gave?”

Official method for calculating NPS

Destination link: Bain—Measuring Your Net Promoter Score

Customer Effort Score (CES)

CES measures how easy or difficult it is to complete a task. For example: “It was easy for me to resolve my issue through customer support.” Responses may be recorded on a 1-to-5 or 1-to-7 scale.

The most common reporting method is the average score:

Average CES = Sum of valid scores ÷ Number of valid responses

Educational example: If the total score from 100 responses on a 1-to-5 scale is 380, the average CES is 3.8.

• Use: Registration, returns, payment, filing a claim, problem resolution, or any process in which friction matters.
• Advantage: Directly reveals process difficulty.
• Limitation: The direction of the scale must be clear. In some questions, a higher score means “easier”; in others, it means “more effort.” Without the exact question wording, the number cannot be interpreted correctly.

Customer Retention Rate

This metric shows what proportion of customers present at the beginning of a period remain through the end of that period.

Retention rate = ((Customers at the end of the period − New customers acquired during the period) ÷ Customers at the beginning of the period) × 100

Educational example: A business starts a period with 1,000 customers, ends with 950, and acquires 150 new customers during the period. Its retention rate is 80%.

• Use: Subscription-based businesses, banking, insurance, and ongoing services.
• Limitation: “Active customer” and the length of the period must be defined precisely. Retention does not explain why customers stay and may be influenced by barriers to exit.

Customer Churn Rate

Churn rate = (Customers lost during the period ÷ Customers at the beginning of the period) × 100

If 80 of the 1,000 customers at the beginning of a month become inactive or cancel according to a clearly defined rule, the illustrative churn rate is 8%.

• Use: An early warning of a declining customer base.
• Limitation: In businesses with infrequent purchases, identifying the exact point of churn is difficult. The usual purchase interval and the behavior of each segment should be considered when defining churn.

Complaint Rate and First Contact Resolution

The complaint rate can be calculated relative to the number of customers, orders, or transactions. For example:

Complaints per 1,000 transactions = (Recorded complaints ÷ Relevant transactions) × 1,000

FCR, or First Contact Resolution = (Issues resolved on first contact ÷ Total eligible issues) × 100

Educational example: If 350 out of 500 eligible support requests are resolved during the first contact, FCR is 70%.

A low complaint rate does not always indicate satisfaction. Difficult access to complaint channels or a lack of confidence in follow-up can also reduce recorded complaints. FCR should likewise be assessed alongside resolution quality and the absence of repeat contact; closing a ticket quickly is not the same as genuinely resolving the issue.

 

CSAT, NPS, or CES: Which One Should You Use?

Metric	What Does It Measure?	Best Time to Measure	Example Use	Interpretation Caution
CSAT	Satisfaction with an interaction or experience	Immediately or shortly after the interaction	Order delivery, branch visit	Depends on the scale and timing
NPS	Willingness to recommend and overall attitude	During periodic measurement or after a meaningful experience	Relationship with a brand or product	Does not explain the reason for the score or actual behavior on its own
CES	Ease of completing a task	Immediately after the process	Returns, registration, issue resolution	The question direction and scale must be clear

Customer Satisfaction Evaluation Methods

The right method depends on the problem, target population, budget, access, and type of decision. The main market research methods used in this area include:

• Online surveys: Fast and suitable for digital customers, but nonresponse bias and unequal access to online channels should be controlled.
• Telephone interviews: Allow questions to be clarified and the sample to be controlled more effectively. Interview length and interviewer training quality are especially important.
• Face-to-face interviews: Appropriate for points of sale, branches, or groups with limited digital access, although costs and interviewer-presence effects are higher.
• Focus groups: Help uncover customers’ language, expectations, and the reasons behind satisfaction or dissatisfaction. Their findings are not suitable for estimating percentages.
• In-depth interviews: Useful for complex experiences such as insurance claims, ending a B2B relationship, or sensitive complaints.
• Complaint analysis: A rich source for identifying process failures, but it captures only the voices of customers who chose to complain.
• Social media comment analysis: Reveals spontaneous feedback; however, social media users are not necessarily representative of the entire customer base.
• Lost-customer interviews: Clarify the real reasons for leaving, competitors’ offers, and the point at which the relationship broke down.
• Mystery shopping: Examines compliance with service standards in real situations; it does not measure the perceptions of all customers.
• Behavioral and transactional data analysis: Connects repeat purchases, cancellations, repeat contact, task-completion time, and process abandonment with attitudinal data.

Quantitative vs. Qualitative Market Research

Approach	Main Question Answered	Typical Output	Example Method
Quantitative market research	How much? In which group? What is the trend?	Percentages, averages, indices, group differences, statistical models	Online, telephone, or face-to-face surveys
Qualitative market research	Why? How? Through what language and logic does the customer interpret the experience?	Patterns, motivations, expectations, pain points, hypotheses	Focus groups, in-depth interviews

Combining methods usually produces a more complete picture. For example, app analytics may show that users drop out at the identity-verification stage; a survey measures the level of dissatisfaction and identifies the groups affected; and in-depth interviews reveal that uncertainty about why information is being requested, together with an unclear error message, is causing users concern.

The Role of Market Research in Measuring Customer Satisfaction

A professional market research project is not limited to asking, “How satisfied are you?” The standard process begins with a business decision and ends with actionable recommendations.

1. Define the business problem: The issue may be rising churn, declining repeat visits, differences in branch performance, or complaints about a process.
2. Set the research objectives: Clarify what will be measured and which decision the findings are intended to support.
3. Identify the target population: Active customers, new buyers, lost customers, complainants, and organizational decision-makers each offer a different perspective.
4. Design the sampling method: The sample should represent the relevant groups as closely as possible and reflect the appropriate shares of branches, channels, products, and segments.
5. Determine the sample size: Sample size depends on the required precision, population variability, subgroup analysis, response rate, and operational constraints. There is no single fixed number suitable for every study.
6. Design and pretest the questionnaire: Wording, scales, order, and questionnaire length should be tested with a small group before the main fieldwork.
7. Collect data and conduct quality control: Interview duration, response patterns, duplicate data, skip logic, interviewer monitoring, and partial back-checks should all be controlled.
8. Analyze the results statistically: In addition to descriptive metrics, group differences and statistically meaningful relationships should be examined.
9. Identify drivers of satisfaction and dissatisfaction: Determine which factors are more closely associated with overall satisfaction, recommendation, or intention to continue the relationship.
10. Present actionable recommendations: Each finding should be translated into a priority, action owner, timeline, and follow-up metric.

[Infographic: Customer satisfaction measurement project stages—Problem definition → Research design → Sampling → Questionnaire design and pretesting → Data collection and quality control → Driver analysis → Action plan → Follow-up measurement]

Using an independent provider of market research services can offer three important advantages: a method tailored to the problem, controlled execution, and more impartial interpretation of the results. Operational independence is especially important when internal units are evaluated using the same metric or respondents may hesitate to express negative opinions directly to brand employees. Outsourcing alone, however, does not guarantee quality. The study’s methodology, sample, questionnaire, quality controls, and limitations should remain transparent and reviewable.

How to Design a Customer Satisfaction Survey

Customer satisfaction questionnaire design does not begin by selecting a few popular questions; it begins with the decision the survey must inform.

• For every question, specify which decision its result will support.
• Keep questions short, clear, and focused on one topic.
• Avoid leading questions such as, “How satisfied are you with our fast and professional service?”
• Use balanced scales with clear labels, and keep them consistent throughout the study.
• Arrange questions from a specific experience to an overall evaluation, or follow the logic of the customer journey.
• Separate overall satisfaction from satisfaction with attributes such as price, speed, quality, and support.
• Include a “Don’t know/Did not experience” option where necessary to avoid forcing meaningless responses.
• Include at least one open-ended question to uncover the reason behind a score or suggestions for improvement.
• Pretest the questionnaire to identify ambiguity, actual completion time, and the performance of skip logic.

Sample Short Customer Satisfaction Survey

This example is designed for an online retailer and should be adapted to the objective, channel, and population of each project.

1. Which product category did your most recent purchase from the store belong to?
2. Overall, how satisfied were you with your most recent shopping experience? (1 = Very dissatisfied to 5 = Very satisfied)—CSAT
3. How satisfied were you with the quality of the product information? (1 to 5)
4. How satisfied were you with the payment process? (1 to 5)
5. How satisfied were you with the delivery time? (1 to 5 / Did not experience)
6. “It was easy for me to complete my purchase on the website.” To what extent do you agree? (1 = Strongly disagree to 5 = Strongly agree)—CES
7. Did you experience a problem with this purchase? (Yes/No) If yes, was it resolved on the first contact?
8. On a scale from 0 to 10, how likely are you to recommend this store to a friend or colleague?—NPS
9. What was the main reason for the score you gave in the previous question?—Open-ended response
10. If we could improve only one aspect of your shopping experience, what would you want it to be?—Open-ended response

Common Mistakes in Customer Satisfaction Measurement

• Using an unsuitable sample: Surveying only highly active customers, app users, or readily available respondents excludes the voices of other groups.
• Making the questionnaire too long: Fatigue leads respondents to abandon the survey or answer in repetitive patterns.
• Measuring at the wrong time: Asking too early, before the experience is complete, or too late reduces recall accuracy.
• Focusing only on the average: The same average can result from two completely different response distributions.
• Ignoring dissatisfied or lost customers: A study limited to customers who remain may create an overly optimistic picture.
• Using one metric for every touchpoint: NPS, CSAT, and CES serve different purposes.
• Failing to analyze group differences: The overall average may hide a poor experience in a particular city, branch, or valuable segment.
• Collecting data without taking action: Repeating surveys without addressing problems erodes trust among customers and employees.
• Interpreting results selectively: Cherry-picking charts or changing calculation rules to make results look better destroys the study’s value.
• Making invalid comparisons across waves: Changes in the scale, fieldwork channel, sample composition, or measurement timing can create an artificial trend.

Customer Satisfaction Study Quality-Control Checklist

• The problem and management decision are clearly defined.
• The target population and respondent eligibility criteria are specified.
• The sampling frame, subgroup shares, and response rate are documented.
• The questionnaire has been pretested and contains no leading or ambiguous questions.
• The timing of measurement is aligned with the relevant touchpoint.
• Metric calculation rules were established before analysis.
• Duplicate responses, unusual completion times, and suspicious patterns have been checked.
• Any weighting and its effect on the findings are reported.
• The study’s limitations and the extent to which findings can be generalized are transparent.
• An action owner and follow-up metric are assigned to every key finding.

How to Analyze Customer Satisfaction Results

Customer satisfaction analysis should begin with describing the data and progress toward explanation and prioritization.

1. Average and response distribution: In addition to the mean, examine the share of each response option and the percentage of highly negative responses.
2. Comparison over time: Compare only periods in which the question, scale, method, and sample composition are sufficiently consistent.
3. Comparison across customer groups: New, long-standing, high-value, infrequent, or complaining customers may have different experiences.
4. Analysis by branch, region, product, or channel: Interpret differences alongside sample size and customer characteristics.
5. Correlation analysis: Examine the relationship between satisfaction with specific attributes and overall satisfaction. Correlation alone does not establish cause and effect.
6. Satisfaction driver analysis: Use regression or other suitable models to identify factors more strongly associated with the primary outcome.
7. Importance–performance gap analysis: Give priority to important factors with weak performance.
8. Open-ended response analysis: Thematic coding, theme frequency, and tone analysis help explain the “why.”
9. Customer segmentation: Identify groups with similar needs, attitudes, or experience patterns so the same action is not prescribed for everyone.
10. Linking attitudinal and behavioral data: Subject to privacy requirements, combine ratings with repeat purchases, repeat contact, cancellations, or task-completion time.

Hypothetical Example: An Online Store and Delivery-Time Problems

Suppose an online store has an overall satisfaction score of 4 out of 5, which appears positive at first glance. A closer analysis, however, presents the following picture:

Factor	Satisfaction out of 5	Hypothetical Relative Importance out of 100	Status
Product quality	4.3	25	Maintain performance
Price	3.8	20	Monitor
Ease of purchase	4.1	15	Maintain performance
Delivery time	2.9	30	Improvement priority
Support responsiveness	3.4	10	Secondary improvement

All figures in this table are hypothetical and provided for educational purposes only. Delivery time has both the lowest performance score and the highest relative importance. Analysis of open-ended responses also reveals themes such as “frequent schedule changes,” “lack of updates,” and “delivery windows that are too broad.” The management message “overall satisfaction is good” is therefore incomplete. The priority action should focus on the reliability of the delivery promise and customer communication.

[Importance–performance chart: Horizontal axis = performance from 1 to 5; vertical axis = relative importance from 0 to 100. Points: product quality 4.3/25, price 3.8/20, ease of purchase 4.1/15, delivery time 2.9/30, support 3.4/10. All data are hypothetical and educational.]

From Data to Action: Turning Results into an Improvement Plan

A good report does more than describe a problem. It connects evidence with an action, an owner, and a follow-up metric.

Identified Problem	Research Evidence	Importance	Recommended Action	Responsible Unit	Follow-Up Metric
Delayed and uncertain delivery times	Satisfaction of 2.9 out of 5 + repeated theme in open-ended responses	High	Review delivery promises and order-status messages	Operations and logistics	On-time delivery, delivery CSAT
Repeat contact for the same issue	Low FCR among complainants	High	Improve resolution guidance and agents’ access permissions	Customer support	FCR, repeat-contact rate
Difficult returns process	Low CES and process abandonment	Medium to high	Simplify the steps and rewrite the messages	Product and customer experience	Returns CES, completion rate
Unclear product information	Frequently repeated theme in open-ended responses	Medium	Standardize specifications and images	Content and commercial teams	Product-information CSAT, return rate

Immediate actions: Fix critical errors, contact customers with unresolved issues, correct misleading messages, and assign a response owner.

Short-term actions: Redesign a high-friction step, train employees, revise the SLA, complete help content, and run a pilot.

Long-term actions: Change infrastructure, redesign the customer journey, revise the service model, or integrate data systems.

[Flowchart: Turning customer feedback into corrective action—Receive feedback → Validate and categorize → Assess frequency and importance → Assign action owner → Run pilot → Monitor metric → Standardize or revise]

To avoid creating a long, unmanageable list, score each action based on its likely impact, urgency, cost, feasibility, and number of customers affected. Then select a limited number of priorities and measure their effect in the next wave.

Example of a Customer Satisfaction Measurement Project

The following study is hypothetical but realistic. All figures are provided solely to explain the method.

Initial problem: A home-appliance repair service company is experiencing a rise in follow-up calls. Managers do not know whether the main issue is technician delays, repair quality, price, or communication.

Research method: A mixed-method design combining support-call analysis, 12 in-depth interviews with customers who had positive and negative experiences, and a telephone survey conducted after service requests were closed.

Population and sample: Customers who received a service during the previous three months. For this educational example, 400 interviews are planned, with controlled shares based on city, appliance type, and type of service. This sample size is not a ready-made prescription for a real project and should be determined according to the analytical objectives.

Metrics: Overall CSAT; satisfaction with appointment timing, technician behavior, cost transparency, and repair quality; service-coordination CES; NPS; FCR; and repeat contact.

Hypothetical findings: Satisfaction with technician behavior is high, but communication about the arrival window receives a lower score. Customers who had to make follow-up calls report weaker CSAT and CES. In open-ended responses, “uncertainty about the appointment time” is mentioned more often than the waiting time itself.

Recommended action: Provide narrower arrival windows, send a reminder before the visit, enable customers to view the technician’s status, and give support staff greater authority to adjust appointment times.

Expected outcome: Fewer follow-up calls, easier service coordination, and higher satisfaction with the process. This outcome should be measured after the pilot by comparing the test group with the baseline and should not be presented in advance as a guaranteed effect.

The study’s value lies not in a single final score, but in showing that customers are not dissatisfied with “a long wait”; they are dissatisfied with “an uncertain wait that requires follow-up.” The distinction changes the appropriate managerial solution.

How Often Should You Measure Customer Satisfaction?

There is no single schedule that suits every business. Measurement frequency should align with the frequency of customer interactions, the speed at which the experience changes, and the organization’s capacity to act.

Measurement Type	Timing	Use	Example Industry
Transactional	Immediately or shortly after an interaction	Evaluating a specific touchpoint	Online shopping, support calls, food delivery, branch visits
Periodic	Monthly, quarterly, or at defined intervals	Assessing the overall relationship and metric trends	Banking, insurance, B2B, subscription services
Continuous	Ongoing feedback collection supported by dashboards and alerts	Managing high-frequency operations	Apps, contact centers, multi-branch retail

For a restaurant, a short transactional survey after an order, combined with a more in-depth periodic study, may be appropriate. A bank may measure major touchpoints such as account opening or request handling transactionally while also running a periodic relationship study. In after-sales service, measurement should occur after customers have had enough time to evaluate repair quality—not at the exact moment the technician leaves.

The key principle is this: do not collect data faster than the organization can analyze it and act on it. Continuous surveys create value only when alert rules, response owners, and a closed-loop customer follow-up process are in place.

Conclusion

Customer satisfaction measurement is not a simple survey or a decorative dashboard score. It is a structured market research process that begins with problem definition and understanding customer needs, continues through sampling, questionnaire design, controlled data collection, and customer data analysis, and must end with trackable action.

CSAT is useful for satisfaction with an interaction, NPS for willingness to recommend, and CES for process ease; however, none provides a complete picture on its own. Combining customer satisfaction surveys with qualitative market research, behavioral data, complaints, and the voice of the customer enables a more accurate diagnosis of the problem.

If your organization plans to design or redesign a satisfaction study, working with an experienced market research team can help define the problem, select the method, design the customer satisfaction questionnaire, conduct fieldwork, manage quality control, and convert results into an action roadmap. A short initial meeting to clarify the management decision, target population, and available data is a practical place to begin.