|
A paper presented by Rodger Gallagher
to the Customer Satisfaction Measurement 2 Conference, held at the
Pan Pacific Hotel, Auckland on the 25 and 26 May 1992.
Are you still measuring the right thing?
Returning to your customers
Have your customers changed?
Are you still market responsive?
RETURNING TO YOUR CUSTOMERS
A - The importance of doing this
Any survey is a snapshot of a moment in time. If everything
stayed the same i.e.: customers didn't change internal processes
in your company remained unchanged and the outside environment was
stable then there would be no need to return to your customers but
everything does change.
In a CSM you are:
1.Finding out what factors are important to your customers.
2. Getting a score on how well you are doing on each of them.
Then you are (presumably) setting up programmes to improve your
scores for the important factors.
What changes?
1. Customers change
The better the service that you supply, the more they demand because
you have raised their expectations. Or the competition have. Their
expectations - of the competition as well as you - may change. Customers
may vary in different locations.
2. The outside environment changes
Competitive companies may arrive, depart, or change the way they
do things. Outside influences may effect the quality of what you
supply - raw materials for a product, the services you can offer,
etc.
3. Internal processes change
Even making a business improvement in your company may quite dramatically
change the way you customers think about you. Improving one thing
may cause another factor to become important, and cause others to
drop off in importance.
In other words, yes it is important to return to your customers
because improvement is an ongoing process. The important questions
are "When" and "How".
B - When?
Different companies have their own preferences as to how often
they should return to their customers,
Some ideas are:
1. After a regular interval
Some companies, e.g.: the First Industry Bank of California,go back
to their customers to update their questionnaires after a fixed
period, in this case every 3 years. If you cant detect the changes
that are taking place this may be the option for you.
2. On an ongoing basis
The Dell Computer Corporation in Austin, Texas, analyze 5,000 -
7,000 daily calls from customers and feed this information back
into their survey system. This is probably the most sensitive check,
but it can be expensive. Ideally it can be built into your measurement
programme (as we will see it later).
3. After any major changes in the marketplace
A pretty obvious pointer. For example, a new competitor or a drastic
price change. Note that these changes in the market place don't
have to occur in your specific sector. Expectations can be modified
(and particularly raised)
4. After internal changes
As we mentioned, as the result of improving something customers
considered important earlier. Once a critical factor is improved,
it may no longer be critical, but something else will certainly
replace it. Often a factor will on1y be critical where it has been
or is a problem, (For example "courtesy" is not appropriate
in the United States surveys because everybody is courteous, unlike
in New Zealand).
5. After a drop in response rates
The importance of maintaining high response rates is well documented.
You can never be sure whether the non-respondents love you or hate
you. That is why a sudden drop in these is a big (drop in information
and you will need to from what has happened to cause it.
6. Indications from your analysis
The particular analysis we use at Telecom lets us know exactly how
much of the variation in overall customer satisfaction is explained
by the combination of our critical factors. If this percentage drops
then we immediately know that something else is influencing customer
satisfaction and we need to find out what it is.
(Obviously combinations of these techniques/intervals may be used).
C - How?
We can divide the methods we use for returning to our customers
into 2 types. Firstly we can "return" to customers on
an ongoing basis which is ideal. It may of course need to be supplemented
with a one-off qualitative study if there are indications of a major
change among our customers. These ongoing methods include:
1. Daily interactions with customers
For example, the calls from customers used by Dell Computers. Or
it may involve a look at the customer complaints file (Although
this may not be a thorough picture) Listening to front line employees
is an important source of customer information. Looking at customer
purchasing behaviour may also be revealing.
2. Regular feedback from interviewers
If respondents are finding some of your questions irrelevant or
are frustrated at not being able to express themselves in certain
areas (because of your structured questionnaire) then they will
tell the interviewers. Unless you or your research company are close
enough to this process to learn about the comments customers make,
this information will be lost.
3. Regular feedback from respondents
You must give respondents an outlet for responses that are not invited
in your questionnaires If you allow interviewers to records "other"
responses, this can be one source, as long as somebody reads them
You can also allow respondents an outlet for any other issues they
would like to raise. This has the advantage of making the respondent
feel they have a valued opinion. There are drawbacks in these methods.
Your questionnaires may take longer to administrate and there will
be increased costs associated with the recording and presentation
of verbatim responses. (Although these costs may be well under the
costs of a qualitative study). To reduce these you may wish to use
these ideas from time to time.
If a more elaborate study is needed it may be necessary to approach
your customers afresh perhaps as you did in the original qualitative
survey. This can satisfy you that you are using the correct wording,
that respondents have a good comprehension of the ideas you are
bringing up, and of course you will be able to identify possible
new factors that have become important since your survey began.
D - What to do about it?
If all the indications exist that you or your customers have changed,
you will need to modify your measurement tool. But a word of warning
here, don't be too quick to change a regular monitor, because it
will lose value for company results over time. If you change the
way a question is worded then it becomes a new question. Even seemingly
trivial wording changes can drastically affect the meaning. So don't
change a question unless you really have to. This doesn't apply
to questions that are obviously being misunderstood by respondents.
These should be improved immediately. Of course you won't be able
to compare these results over time either, but then as the old ones
(which were misunderstood) are useless then it doesn't matter.
Some other suggestions
If you drop a question out make sure it can easily be added
again at some future date if necessary. And it may well be added
back, as things can drop in and out of importance, Maybe you will
have some that you pop in every six months or so.
Don't add so many questions that your survey becomes too long.
Be economic.
If your questionnaire becomes too long, your response rate could
drop. If you put a question back in, that has been used in the past,
make sure you use the exact wording.
|
