Patient Experience, Guide

How to measure Patient Satisfaction?

Processes, services to audit, distribution channels and data analysis for an effective measurement system.

Measuring patient satisfaction is key to continuously improving healthcare quality. With RateNow, data is collected and analysed in real time from a single platform, enabling all levels (strategic, tactical and operational) to make actionable decisions. Measurement is continuous, segmented and multichannel, combining indicators such as NPS and CSAT, multichannel distribution and analytics technologies to process large volumes of feedback.

Riccardo Begelle, 2025

21 min

The first step towards continuous improvement of healthcare quality is having a good patient satisfaction measurement system. Measuring it is a necessary condition for improving it systematically.

A good measurement system enables staff, managers and heads of service to identify which aspects of care do not meet patient expectations and allocate resources where they have the most impact. It is the most powerful tool quality teams have for orientating the organisation towards what truly matters to patients.

Patient Satisfaction: What Are We Measuring Exactly?

Before designing any measurement system, it is essential to clarify what is meant by patient satisfaction and what distinguishes it from other adjacent concepts that are often confused in practice.

Patient satisfaction is an outcome indicator: it measures whether the patient considers that the service received has met their prior expectations. It is a global, eminently subjective assessment that reflects the gap between what the patient expected and what they perceived at the end of their experience. In Donabedian's taxonomy, it is an outcome indicator, not a process indicator.

Concept	What it measures	Typical tool
Patient satisfaction	Whether patient expectations were met with the service received (outcome).	NPS, CSAT, Likert scales
Patient experience	How the care process was: treatment, communication, coordination, information (process).	PREM (PPE, IEXPAC, EUROPEP)
Perceived quality	Patient's technical assessment of the effectiveness of the treatment.	Specific clinical questionnaires

Patient satisfaction and patient experience are complementary but distinct metrics

Satisfaction answers the question 'Was the patient happy?' Experience answers 'How was the care process?' A hospital can have patients satisfied with the clinical outcome but with a poor care experience, or vice versa. Measuring them separately with the right tools is what enables precise action.

Who Should Measure Patient Satisfaction?

According to the research What Matters to Patients? (King's College London), hospital management and central services must use the collected data to support the management and improvement of frontline services.

Level	Actor	Information need
Strategic	Management and Senior Leadership	Global satisfaction indices (NPS, CSAT) and quarterly comparisons between services and sector benchmarks.
Tactical	Quality teams	Detailed monthly reports by service. Identification of areas with scores below the acceptable threshold.
Operational	Middle managers and clinical teams	Weekly reports for their own services. Textual comments from dissatisfied patients.

Linking clinical services and feedback systems

Feedback systems must be directly linked to clinical services so that professionals identify with the results. A whole-hospital satisfaction report does not motivate action: a report for the cardiology unit with comments from its own patients does.

What Do You Need to Measure Patient Satisfaction?

An effective measurement system is not just a survey: it requires a minimum infrastructure to ensure that data is collected, analysed and integrated into organisational decision-making. The four essential resources are:

Qualified staff: With exclusive responsibility for overseeing and coordinating the programme. Feedback without a manager to handle it does not generate improvements.
Dedicated time: For those responsible for each service to periodically review scores and lead concrete action plans.
Budget: For staff training, survey design and updating, and hiring methodology experts if necessary.
Dedicated technology platform: That automates survey distribution, collects responses in a centralised repository and generates dashboards segmented by service and role.

What Characteristics Should a Good Satisfaction Measurement System Have?

Patient satisfaction information only has operational value when it meets these five characteristics:

Characteristic	Why it is essential
Continuous	Satisfaction is fluid and seasonal. Peak activity periods, bank holiday shifts or staff changes affect scores. Annual or semi-annual surveys do not detect these variations.
Segmented	Data must be organised in the same way that clinical teams are organised. The head of the trauma unit needs their patients' scores, not a global hospital average.
Real-time or close to the event	Patient satisfaction and recall degrade rapidly over time. The most reliable surveys are administered within the first 24–48 hours after the interaction.
Easy to complete	Any friction in the process dramatically reduces the response rate. The optimal length is under 5 minutes. Long or hard-to-access questionnaires generate selection biases.
Quantitative and qualitative	Numerical scores (NPS, CSAT) locate the problem and enable comparisons. Open-ended questions explain the reasons behind the numbers. Both are necessary to act with precision.

The Four-Step Measurement Process

The "four-step measurement process" is a structured approach to effective measurement, establishing a link between data collection, analysis, enquiry and reporting patterns.

""Following the four steps ensures that all collected data is used to make better decisions that ultimately improve services."
— NHS Patient Experience Book, 2017

Step 1: Which Services to Measure?

The first step is to identify which services will be audited and in what order. The standard recommendation is to start with the highest-volume services and gradually expand:

Phase	Priority services	Justification
Phase 1 (launch)	Outpatient consultations, Inpatient, A&E, Day surgery	Highest volume of interactions. The impact of improvements is immediately visible in scores.
Phase 2 (expansion)	Radiology, Pharmacy, Rehabilitation, Oncology, Laboratory	Services with specific characteristics where contextualised feedback adds differential value.
Full programme	Up to 30 hospital processes	Full coverage of the patient pathway, including support processes.

The most frequent mistake is wanting to audit everything from day one. A programme that starts with two or three key services, works well and builds trust in teams is far more sustainable than a massive rollout that collapses under its own complexity.

Step 2: How to Design a Satisfaction Survey?

Survey design directly determines data quality and response rates. A well-designed survey must meet these conditions:

Cover the relevant touchpoints: Use the patient journey map to identify the moments of greatest impact on satisfaction: appointment, reception, consultation, waiting, discharge, follow-up. It is not necessary to cover all touchpoints in a single survey; it is better to focus on the most relevant ones for each service.
Ask about both relational and functional aspects: Research by King's College London (2011) shows that patients value relational factors (feeling listened to, respected, informed) and functional factors (waiting times, accessibility, cleanliness) equally.
Include a recommendation indicator (NPS): The Net Promoter Score, which asks 'Would you recommend this service?', is the indicator with the greatest predictive power for patient loyalty and the most comparable with external benchmarks. An increasing number of public institutions are incorporating it as a central indicator.
Include a point-in-time satisfaction indicator (CSAT): The direct satisfaction question ('How satisfied are you with the care received?') complements NPS with a more granular assessment of a specific interaction. The recommended scale is 5 points.
Add at least one open-ended question: A question such as 'Is there anything we could have done better?' generates the most valuable qualitative comments. In 2026, the analysis of these responses is carried out with NLP (Natural Language Processing), which automatically classifies topics and sentiments in real time.
Adapt length to the channel: A questionnaire sent by SMS should have between 3 and 7 questions. An on-site feedback terminal survey should be even shorter: 2 to 5 questions. Long surveys dramatically reduce response rates and generate lower-quality data.

Satisfaction Indicators: NPS and CSAT

The two most widely used indicators for measuring patient satisfaction are the Net Promoter Score (NPS) and the Customer Satisfaction Score (CSAT). They are complementary and measure different aspects:

Indicator	What it measures	Typical question	Calculation
NPS	Loyalty and likelihood to recommend.	'How likely are you to recommend this service?' (0–10)	% Promoters (9–10) − % Detractors (0–6)
CSAT	Point-in-time satisfaction with a specific interaction or service.	'How satisfied are you with the care received?' (1–5)	% responses 4 or 5 over total

"12x more likely to be NPS Promoters are patients who feel treated as individuals compared to those who feel treated as data. (NRC Health, 2025)

Step 3: How to Distribute Surveys?

The choice of distribution channel determines both the response rate and the representativeness of the sample. The most effective programmes combine several channels adapted to the type of service and patient profile.

Channel	Response rate	Best use in healthcare
SMS	15–30%	Inpatient, day surgery, high engagement. The channel with the highest response rate in healthcare.
Email	10–20%	Follow-up services, chronic patients with digital contact history. Lower rate but very low marginal cost.
Smiley feedback Terminals	10–20%	Outpatient consultations, A&E, Primary care. Ideal for high frequency of in-person patients.
QR	~5%	Only if actively requested from the patient. Systematic bias towards younger, digitally literate users.
Telephone	~70%	Specific projects or populations with low digital adoption. High cost per survey.
Paper	Low	In decline due to the operational cost of data entry and low participation rate.

The recommended combination for most hospital organisations is SMS for high-engagement services (inpatient, surgery) and on-site terminals for high-traffic transactional services (outpatient consultations, A&E). Email acts as a backup channel.

A critical aspect is timing: surveys sent within the first 2 hours after discharge or interaction generate significantly higher actionability rates than those sent days later (Mobile Engagement Report 2025). Satisfaction and recall are more precise and less filtered by cognitive dissonance when collected close to the event.

Sampling: Who to Survey?

There are two sampling strategies:

Universal sampling: The survey is offered to all patients who have used the service in the period. This is the recommended standard for services with moderate volume. It maximises representativeness and eliminates selection bias.
Stratified random sampling: A representative sample is selected when patient volume is very high and analytical resources are limited. This is the method used by HCAHPS in the USA. It requires careful statistical design to ensure representativeness by sex, age and type of admission.

The general recommendation is universal sampling when volumes allow, as it provides more granular data and enables segmentation by clinical unit, shift or professional.

AI in Patient Satisfaction Analysis

Measurement platforms such as RateNow apply artificial intelligence in the analysis and action phases of the satisfaction cycle:

NLP on open-ended questions: Natural Language Processing automatically classifies qualitative responses by topic, sentiment (positive, negative, neutral) and urgency. It enables the analysis of thousands of comments in real time without manual review and the detection of emerging patterns such as recurring complaints about a specific service.
Predictive satisfaction analysis: Machine learning models identify correlations between operational variables (waiting times, staff changes, service reorganisations) and changes in satisfaction scores, enabling proactive interventions before the problem becomes systemic.
Automatic alerts and loop closure: When a patient gives a low score or includes a complaint in their comment, the system automatically generates an alert assigned to the service manager with a defined response SLA. This enables immediate contact with the dissatisfied patient and closure of the feedback loop.
Role-personalised dashboards: The same satisfaction database generates different reports depending on the organisational level: a global NPS for management, a breakdown by dimensions for the quality team and anonymised individual comments for the head of service.

The quantitative statistical analysis (means, distributions, trends) can be fully automated. For AI-driven qualitative analysis, the recommendation is that the final interpretation and action decision should remain in the hands of professionals, using AI as a prioritisation tool, not as a substitute for clinical judgement.

Step 4: How to Analyse and Distribute Information?

Data collection only has value if it translates into action. The principles of effective analysis are:

Centralise in a single repository: All satisfaction information must be in a single system accessible to all relevant stakeholders, with differentiated access permissions by level and service.
Differentiated analysis frequency by level: Medical services and unit heads: weekly review of their patients' scores and comments. Quality team: monthly analysis with trend comparisons and alerts. Management: quarterly dashboard with key indicators and evolution.
Funnel approach — from general to specific: Analysis starts from the hospital's global NPS or CSAT → breaks down by service → identifies the service with the greatest drop → analyses the specific dimensions (waiting, communication, friendliness) → reads the qualitative comments from patients of that service. This flow prevents teams from getting lost in aggregated data that does not enable action.
From quantitative to qualitative: Numerical indicators locate the problem and its magnitude. Open-ended comments explain why the problem exists. Never close an analysis cycle without reading a representative sample of comments from the worst-scoring segment.
Internal and external benchmarking: Comparing the performance of each service with the hospital average (internal benchmarking) and with the sector (external benchmarking) is what gives context to raw data. An NPS of 55 can be excellent or mediocre depending on the type of service and the country.
Integrate satisfaction into team meetings: Organisations with the best satisfaction results systematically include feedback data in their regular meetings. At management level: a specific chapter in the quarterly Management Committee. At operational level: weekly review of scores with the clinical team.

When Should You Stop Measuring?

""The answer is 'never'. If you are achieving your overall satisfaction goal, you can always work on improving the relatively lower-rated areas. What gets measured gets improved. What stops being measured tends to deteriorate."
— NHS Patient Experience Book, 2017

Continuous measurement also has a valuable secondary effect: it keeps the clinical and managerial team's attention high on patient satisfaction as a strategic indicator. When measurement is suspended, satisfaction loses visibility on the organisational agenda and tends to deteriorate.

Every hospital or clinic has specific needs and it is important to have experts to make the right decisions, accumulate experience and prevent changes from generating tensions in complex organisations. Implementing satisfaction measurement systems can be delicate for both the organisation and patients.

At RateNow we specialise in reducing complexity and focus on fast, flawless implementations. If you have a patient satisfaction measurement project in mind for your organisation: contact us.

Bibliography

1. King's College London: What Matters to Patients? DH & NHS Institute for Innovation & Improvement, 2011.

2. NHS / DH: The Patient Experience Book. Institute for Innovation and Improvement, 2017.

3. Anzolo Med: HCAHPS 2.0: Navigating the 2025–2026 Patient Experience Survey Changes. January 2026.

4. NRC Health: Patients treated as individuals are 12x more likely to be Promoters (NPS). 2025.

5. Press Ganey: State of Patient Experience 2025. July 2025.

6. Berwick DM, Nolan TW, Whittington J: The Triple Aim: care, health, and cost. Health Affairs. 2008.

What is the difference between patient satisfaction and patient experience?

Satisfaction (CSAT) is an outcome indicator: it measures whether the patient was happy with a specific interaction. Patient experience (PREM) is a process indicator (Donabedian's taxonomy): it measures how care was delivered throughout the entire care pathway, including emotional and relational dimensions that satisfaction does not capture.

What indicators are used to measure patient satisfaction?

The two most widely used are the Net Promoter Score (NPS), which measures the likelihood to recommend on a scale of 0 to 10, and the Customer Satisfaction Score (CSAT), which measures point-in-time satisfaction with an interaction on a scale of 1 to 5. These are complemented by open-ended questions to capture qualitative feedback. HCAHPS in the USA is the national standard that includes both satisfaction and experience dimensions and has a direct impact on Medicare reimbursement.

How often should patient satisfaction be measured?

Measurement must be continuous, not one-off. Annual surveys are useful for normative comparisons but do not enable real-time problem detection or action. Analysis frequency should vary by level: weekly for heads of service, monthly for quality teams, quarterly for management.

What is HCAHPS and why is it relevant outside the USA?

HCAHPS (Hospital Consumer Assessment of Healthcare Providers and Systems) is the US national standard for measuring the patient's perspective on hospital care, mandatory for hospitals billing Medicare. In 2025 it expanded from 29 to 32 questions and incorporated electronic administration for the first time. It is globally relevant because its dimensions (communication with physicians and nurses, discharge coordination, environment) represent the international scientific consensus on which aspects of care have the greatest impact on satisfaction.

When should you stop measuring patient satisfaction?

Never. Continuous measurement has a dual value: it provides data for improvement and maintains the organisational focus on patient satisfaction as a strategic indicator. When measurement is suspended, satisfaction tends to lose visibility on the management agenda and to deteriorate.

Tell us about your project and get:

Best practices applied to your industry

Surveys tailored to your needs

Demo of our analytics platform

A customized quotation

ⓘ No commitment

Patient Experience