Ten Requirements of Successful Customer Satisfaction Surveys
Customer loyalty initiatives have succeeded and failed for a variety of
reasons. The customer satisfaction field has developed a great deal over the
past decade, and practitioners now agree on a number of factors necessary
for a customer loyalty initiative to have a bottom-line impact. Following is
a brief summary of these factors, each of which should be addressed in your
customer satisfaction research.
1. Set the bar high enough. Customer satisfaction has a direct
impact on customer retention and share of wallet, but the relationship is
not linear. It is curvilinear. In competitive markets, a somewhat satisfied
customer is rarely any more loyal than a somewhat dissatisfied customer;
only highly satisfied customers are loyal. Organizations that do not
understand how high to set their customer satisfaction bar will waste a
great deal of money improving their service to a level that has no
bottom-line impact. This means you need to have some type of ďbottom-lineĒ
measure in your survey (anticipated loyalty, percent of purchases, primary
supplier, account balances) as well as an overall satisfaction measure, and
then look at the relationship between the two.
2. Make sure youíre talking to the right customers. Probably the
most important aspect of any customer survey is selecting the right people
to talk with. First, focus on your most important customers; consider the
80-20 rule. Many organizations segment there customers based on sales or
sales potential. Then they survey a larger percentage of major customers,
and a smaller percentage of their smaller customers. Second, identify the
decision-makers; end-users are not always decision-makers. Third, customer
relationships are complex, and in business-to-business situations you may
need to talk with different customers about different aspects of the
relationship. The customer who answers about ordering and delivery may not
be able to tell you about billing practices. Fourth, factor your customer
segments into your research design. And finally, as in all research, be
demanding about response rates. Sample sizes are irrelevant; response rates
among the right customer groups is what counts.
3. Donít focus too much on problem resolution. Research
demonstrating that a dissatisfied customer tells an average of 10-15 people
(depending on the industry) about their negative experience led to a focus
on problem resolution as a means to build loyalty. But loyalty does not come
from having problems resolved; it results from being highly satisfied with
an offering. Most customer satisfaction surveys query customers about
problems and problem resolution; thatís fine, just donít use problem
information as a primary driver in setting improvement priorities.
4. Capture competitive information. Whenever possible, competitor
information should be captured because this is the best way to establish
improvement priorities. If your customers use more than one provider, ask
them to rate both. If not, you may need to conduct a market (as opposed to a
customer) survey. Because capturing competitive information can be
expensive, many organizations conduct a larger baseline survey once every
1-3 years with competitive information, and then conduct smaller tracking
surveys without competitor questions on a more frequent basis.
5. Donít ask customers whatís most important to them; figure it out
from the data. Customers do a poor job of telling us what is really most
important to them. The best way to understand the relative impact of various
factors on a customerís satisfaction and loyalty, is to use an inferred
statistical analysis approach. (There are several analytic techniques used
to determine key drivers of satisfaction, including regression and
6. Use both competitive performance and derived importance ratings to
set priorities. If you capture competitive information, divide the
satisfaction ratings into three buckets: those where you do significantly
better than the competition, those where youíre at par, and those where you
perform below the competition. Use statistical analysis to determine the
relative importance of each factor to the customer, and divide these
importance scores into three groups: high, medium and low. To set
priorities, simply create a three-by-three matrix of competitive performance
by importance. Your first priority is obvious: important areas where you
perform below the competition. In addition, you should have an approach to
the improvement process in place before you conduct your survey to insure
that youíll be able to act quickly on improvement priorities. This does not
mean that you need a formal Quality organization; it simply means that you
have an agreed upon approach to problem solving and action planning. Most
customer satisfaction research consultants can also provide assistance in
the improvement process.
7. Consider asking some informational questions. The clothing
retailer Sy Syms is correct when he says that an educated customer is his
best customer. Understanding how to use a companyís product or service has a
very strong bearing on satisfaction with that product or service. Mercedes
owners who had the operations of their new vehicle explained to them when
they picked it up reported fewer engine problems over the next two years.
Cellular phone customers who understand all the options on their phones are
less likely to switch carriers. Chinese food customers who know the food is
made without MSG are more loyal. Examples cross virtually every industry.
Consider testing your customerís knowledge of key features you provide, and
use any low information scores as an indication that you need to communicate
better with your customers.
8. Use demographics information to help identify root causes.
Traditional cross-tabs can offer a great deal of insight into satisfaction
ratings. Are there differences in satisfaction with delivery performance by
geographic location? Do retail customers think they receive better service
on certain days of the week or times of the day? Is satisfaction higher at
branches with longer tenured managers? To avoid unnecessarily adding length
to your questionnaire, attach key respondent information to your initial
customer list whenever possible. If youíre conducting a longer baseline
followed by shorter tracking surveys, you may only need to add the
demographics questions to your baseline. Just remember that having this
information is going to be extremely important when you get to the problem
9. Use follow-up qualitative customer research to fine-tune
improvement plans. In customer satisfaction, employees with customer
contact can usually provide all the information you need to design your
questionnaire, thus eliminating the need for initial qualitative research
with customers. But really understanding how to delight customers is a
little like pealing the onion. The quantitative survey gives specific
direction on improvement priorities, but can rarely provide all the details.
Subsequent qualitative research is often needed to help define and test
recommendations. In addition, in industries with a small number of large
customers, itís highly recommended that your President send a follow-up
letter to customers when the survey is completed. This is a short
communication that simply says: thank you for participating in our survey,
this is what we heard, and this is what we intend to do about it.
10. Create focus and donít try to be all things to all people. The
biggest mistake in customer satisfaction is the temptation to try to do too
much in an effort to delight customers and exceed their expectations.
Rather, successful companies understand how to set priorities, to focus on
the key issues, and to align their organization around key success factors.
You are most likely to delight your customers by focusing on a small number
of very important issues that differentiate you from the competition. Such
focus accomplishes two things. First, it forces you to focus scarce
resources on the most important things, thus increasing your likelihood of
success. And second, it allows you to fully develop your improvement process
before you attack additional problems.
All these issues should be factored into your customer satisfaction
survey. A well designed customer satisfaction survey can be an invaluable
tool to management, but a poorly designed program will likely result in
wasted resources and demotivated employees.