Cabinet Office

Cabinet Office sites

Related Links

• National Consumer Council
• 10 Downing Street
• Strategy Unit
• Delivery Unit

Collecting Customer Data –
A Framework For Government Departments

OPSR has established 7 principles of customer feedback:

Customer satisfaction data is gathered in many different ways in public services. In order for it to be useful, all these principles should be together or we risk basing services on assumption and expertise as we have in the past. It is only on the basis of excellent data that we can move forward to respond to the new policy context.

• Listen And Respond
• Local Level
• National Comparability
• Segmenting Customers
• Drivers Of Satisfaction
• Effectively
• Over Time

(1)We should listen and respond to customers.

By customers we mean the people who use the service. In some cases there may be more than one customer group, for example both parents and children can be defined as customers of primary schools.

As a minimum, a customer survey should include:

An overall rating

• Frequently a single measure of overall customer satisfaction, although other overall rating questions may be appropriate.

Ratings of specific aspects of service delivery

• Service attributes will include measures of delivery, timeliness, information, professionalism and staff attitude, which have been found to be generic drivers of satisfaction. Further service attributes will depend on the service provided.
• These rating questions are used in key driver analysis to derive which aspects of service delivery have greatest impact on a positive customer experience (see (5) below)

Customer expectations

• What do customers expect of the service provided? By asking respondents to relate their actual experience of services in conjunction with their levels of satisfaction, we can derive what level of service standard is acceptable to customers.

(2) At the local level

Data needs to be collected at the local level. For data to be actionable, we need to be able to understand the customer experience at the relevant local management unit, such as individual school, college, basic command unit, hospital etc.

Data should be collected consistently by individual local management unit so that results can be compared between peers.

(3) Ensuring national comparability

Being able to make comparisons in a national context between different geographies, demographic groups, providers or management units of similar services is crucial to establishing and maintaining standards and accountability. Data needs to be collected locally in such a way that it can be aggregated to give the national picture.

(4) Segmenting customers

As customers are not a homogenous group, it is important to be able to analyse customer survey data by different customer group. Customer segmentation needs to reflect the true spectrum of customer groups that are served rather than the current internal structures of service deliverers. Customer segmentation will differ from service to service; some examples of ways in which customers can be segmented include:

• Demography
• Geography
• Frequency of service use
• Motivation for service use
• Customer type

(5) Seeking to understand the factors, which drive satisfaction

We need to know which factors have greatest impact on customers’ overall experience. To do this, we need to

• Ask an overall rating question
• Ask rating questions of all aspects of service delivery (see 1 above)
• Carry out secondary statistical analyses to derive the key drivers

(6) Reliably and cost effectively

• Methodologies consistent at the local level in terms of
  1. data collection method, questions asked, question order, sampling procedures, time of fieldwork, analysis and reporting
• Surveys are inclusive as possible
  1. Particular effort made to include hard to reach groups
  2. Techniques used to ensure as high a survey response rate as possible and analysis carried out among to understand impact of possible non-response biases
• Cost benefit analysis carried out
  1. At what point does increased survey cost outweigh further improvements to data quality?

(7) Over time

Methodologies need to be consistent over time to give reliable trend data.

The decision on how regularly to survey customers is a trade-off between the need for frequent performance data which reflect actual change in the customer experience and the financial and administrative burdens.

Many surveys are carried out twice a year or more frequently, and increasingly customer research is carried out continuously throughout the year, with quarterly reporting of data.