Measuring customer satisfaction: methods, good questions and what to do with the result

Why do you measure customer satisfaction?

Not every dissatisfied customer complains. Many say nothing, choose someone else next time and at most tell their colleagues. A measurement brings that quiet signal to the surface before it costs you work.

For a quality management system such as ISO 9001 it is also a core topic: you follow how customers experience your work and use that to improve. The result is therefore not an end in itself, but input for objectives, actions and the management review.

Which methods are there?

No single method tells the whole story. Combine two or three that fit the number of customers you have. With a few large clients a conversation works better than a survey, with hundreds of customers it is the other way round.

With business customers, the person who decides differs from the person who experiences your work. The buyer looks at price and agreements, the project manager at planning and the user at quality. So ask several people per customer, otherwise you miss part of the picture. The methods you can combine:

• Survey: a short questionnaire per project, delivery or year, with a fixed scale
• NPS: one question on whether customers would recommend you, with a follow-up question on the reason
• Customer conversation: an evaluation after completion or an annual talk with your most important customers
• Complaint analysis: what customers tell you unprompted, sorted by subject and trend

NPS: one question, with a follow-up

The Net Promoter Score asks: how likely are you to recommend us, on a scale of 0 to 10? Anyone giving 9 or 10 is a promoter, 7 or 8 is passive and 0 to 6 is a detractor. The score is the percentage of promoters minus the percentage of detractors.

The strength is the simplicity and the trend over time. The weakness is that the number does not say why. So always ask an open follow-up question about the main reason. If you have only a limited number of customers, do not look at the percentage but at each answer individually, because a single customer already shifts the score considerably.

What are good questions?

Good questions are short, cover one subject and do not steer. So do not ask "Are you happy with our fast and skilled service?", because that is two things in one. Use the same scale throughout, for example 1 to 5, and finish with an open question. Adapt the subjects to your work. A few examples:

• How satisfied are you with the quality of the work delivered?
• How satisfied are you with our planning and keeping of agreements?
• How satisfied are you with our communication and availability?
• How satisfied are you with the safety and conduct of our people on site?
• What is the first thing you think we should improve?

How often do you measure?

There are two kinds of measurements. A transactional measurement follows a concrete moment, such as the completion of a project or the handling of a complaint. A relational measurement looks at the whole cooperation and you do it periodically, for example yearly or twice a year.

Choose a rhythm you can sustain. A short measurement you repeat every year is worth more than an extensive study that stops after one round. That way you can see trends and measure the effects of improvements.

How do you get a higher response?

People fill in a survey when it takes little effort and when it makes sense. Give a reason why you ask and show what you did with earlier answers. Ask the regular contact person to send the invitation, because a personal request works better than an anonymous email. This is how you keep the threshold low:

• Keep the questionnaire short: five good questions rather than twenty
• Send at the right moment, shortly after completion or finishing
• Make sure filling it in works on a phone
• Send one reminder, not a series
• Call important customers who do not respond yourself

How to use the result

A measurement without follow-up is worse than no measurement, because customers notice that nothing happens. Call a customer who gives a low score back within a few days to hear what is going on. Turn recurring points into actions, with an owner and deadline.

An example: three clients indicate that they only hear late what extra work costs. You agree that the foreman confirms extra work in writing within a day, and at the next measurement you check whether that point scores better. That way a number becomes an improvement the customer sees.

Set a target, such as a minimum average score or an NPS level you find realistic yourself, and follow the trend per quarter. Combine it with your complaints and quality reports, because together they give the most honest picture. Discuss the outcome in the management review and let customers know what you did with it.

Common mistakes when measuring customer satisfaction

The biggest mistake is measuring without doing anything with the result. A second mistake is only asking customers who are satisfied anyway. Then the average looks good, but exactly the signals you need to learn from are missing.

Also watch out for an average that hides everything. A score of 4 can consist of satisfied customers and one customer about to leave. So look at individual answers and at customers with high turnover or high risk.

How BrydgeOne supports this

In Satisfaction you survey customer and employee satisfaction and analyse the answers. You record complaints and deviations in Quality reports. With Objectives & KPIs you steer on measurable goals and trends, and Management review compiles the management review from live data.