The approach to service segmentation suggested here follows a three-step
process:
1 Identify the key components of customer service as seen by customers
themselves.
2 Establish the relative importance of those service components to customers.
3 Identify ‘clusters’ of customers according to similarity of service preferences.
1 Identifying the key components of customer service
A common failing in business is to assume that ‘we know what our customers
want’. However, the truth is that it is easy to become divorced from the reality of
the marketplace when management is consumed with the day-to-day pressures of
running a business. How should we know which aspects of service are most highly
rated by the customer? Given the complexity of the market that the typical company
serves how might it better understand the segmentation of those markets in
terms of service requirements? What does it take for a company to become the
supplier of choice?
Clearly it is important to develop an understanding of the service needs of customers
through detailed research.
The first step in research of this type is to identify the key sources of influence
upon the purchase decision. If, for example, we are selling components to
a manufacturer, who will make the decision on the choice of supplier? This is not
always an easy question to answer as in many cases there will be several people
involved. The purchasing manager of the company to which we are selling may
only be acting as an agent for others within the firm. In other cases his or her influence
will be much greater. Alternatively if we are manufacturing products for sale
through retail outlets, is the decision to stock made centrally or by individual store
managers? The answers can often be supplied by the sales force. The sales representative
should know from experience who the decision makers are.
Given that a clear indication of the source of decision-making power can be
gained, the customer service researcher at least knows who to research. The
question remains as to which elements of the vendor’s total marketing offering
have what effect upon the purchase decision.
Ideally once the decision-making unit in a specific market has been identified,
an initial, small-scale research programme should be undertaken based upon
personal interviews with a representative sample of buyers. The purpose of these
interviews is to elicit, in the language of the customers, firstly, the importance they
attach to customer service vis-à-vis the other marketing mix elements such as
price, product quality, promotion, etc., and secondly, the specific importance they
attach to the individual components of customer service.
The importance of this initial step in measuring customer service is that relevant
and meaningful measures of customer service are generated by the customers
themselves. Once these dimensions are defined we can identify the relative importance
of each one and the extent to which different types of customer are prepared
to trade off one aspect of service for another.
LOGISTICS AND CUSTOMER VALUE 41
2 Establishing the relative importance of customer service components
One of the simplest ways of discovering the importance a customer attaches to
each element of customer service is to take the components generated by means
of the process described in step 1 and to ask a representative sample of customers
to rank order them from the ‘most important’ to the ‘least important’. In practice
this is difficult, particularly with a large number of components, and would not give
any insight into the relative importance of each element. Alternatively a form of
rating scale could be used. For example, the respondents could be asked to place
a weight from 1 to 10 against each component according to how much importance
they attached to each element. The problem here is that respondents will tend to
rate most of the components as highly important, especially since those components
were generated on the grounds of importance to customers in the first place.
A partial solution is to ask the respondent to allocate a total of 100 points amongst
all the elements listed, according to perceived importance. However, this is a fairly
daunting task for the respondent and can often result in an arbitrary allocation.
Fortunately a relatively recent innovation in consumer research technology now
enables us to evaluate very simply the implicit importance that a customer attaches
to the separate elements of customer service. The technique is based around the
concept of trade-off and can best be illustrated by an example from everyday life.
In considering, say, the purchase of a new car we might desire specific attributes,
e.g. performance in terms of speed and acceleration, economy in terms of petrol
consumption, size in terms of passenger and luggage capacity and, of course, low
price. However, it is unlikely that any one car will meet all of these requirements so
we are forced to trade off one or more of these attributes against the others.
The same is true of the customer faced with alternative options of distribution
service. The buyer might be prepared to sacrifice a day or two of lead time in order
to gain delivery reliability, or to trade off order completeness against improvements
in order entry, etc. Essentially the trade-off technique works by presenting the
respondent with feasible combinations of customer service elements and asking for
a rank order of preference for those combinations. Computer analysis then determines
the implicit importance attached by the respondent to each service element.9
Whatever technique is used it is important to understand which are the ‘qualifiers’
and which are the ‘order winners’ amongst the different customer groups. This
understanding must then drive the design of the supply chain processes that will
enable success in the marketplace.
3 Identifying customer service segments
Now that we have determined the importance attached by different respondents
to each of the service attributes previously identified, the final step is to see if any
similarities of preference emerge. If one group of respondents, for example, has a
clearly distinct set of priorities from another then it would be reasonable to think of
them both as different service segments.
How can these customer service segments be identified? One technique that
has been successfully used in this connection is cluster analysis. Cluster analysis
is a computer-based method for looking across a set of data and seeking
42 LOGISTICS & SUPPLY CHAIN MANAGEMENT
to ‘match’ respondents across as many dimensions as possible. Thus if two
respondents completed the step 2 trade-off analysis in a similar way their importance
scores on the various service dimensions would be similar and hence the
cluster analysis would assign them to the same group.
One study in an industrial market suggested that the traditional way of segmenting
customers according to ‘Standard Industrial Classification’ (SIC) had
little relevance to purchasing behaviour. The classic categorisation of customers
according to industry sector did not correlate with the attributes they sought
from suppliers. Instead it seemed that some companies were very time-sensitive
in terms of delivery reliability – a ‘just-in-time’ segment – regardless of the industry
they were in. In the same way there was a very clear ‘price’ segment, which
also cut across conventional industrial classifications. A further segment was much
more responsive to a ‘relationship’ approach, valuing technical support and close
supplier liaison much more highly. As a result of this research the supplier was
better able to focus its marketing efforts and to re-engineer its supply chain strategy
to achieve a better match with customer requirements.
The challenge to logistics management is to create appropriate supply chain
solutions to meet the needs of these different value segments. More than likely
there will be the need for multiple supply chain solutions since ‘one size will not fit
all’. This issue will be dealt with in detail in Chapter 5 where the concept of supply
chain agility is discussed.