Recent moves toward a more competitive
retail environment, as well
as increased performance expectations
from shareholders, have focused
utilities on improving the customer experience
and, to some extent, on improving
customer profitability. However, many
utilities continue to have only a very limited
understanding of the unique financial
costs and benefits associated with each
customer. While regulatory requirements
and the societal obligation to provide
basic utility services will certainly continue,
even for unprofitable customers, we
believe utilities need to better understand
profitability at the customer level in order
to improve the customer experience and
contribute to the bottom line.

This article begins with a discussion
of the benefits of utility customer profitability
analysis, then outlines a unique
approach to modeling profitability for
individual customers and customer segments
and provides an example of some
successful profitability analysis results.
All utilities should consider performing
this type of analysis to support their
efforts to improve the customer experience,
increase revenues, reduce costs and
achieve profitable growth.

Value of Customer Profitability Analysis

Customer profitability analysis is most
effective when combined with both demographic
and customer satisfaction data.
The result is a powerful tool that utilities
can use to create new value in four primary

Know What Is Driving Profit and Loss
This seems very basic, but utilities often
don’t understand what drives their profitability.
As a result, they have difficulty
answering some important questions, such
as: Are sales to residential customers during
weekday peak demand periods resulting
in significant losses? How do repeat
calls from certain customers to address
basic billing questions affect the profitability
of those customers? Should programs
to drive customers toward online billing
programs be expanded, or do those programs
result in a financial loss? Customer
profitability analysis can help answer
these and other important questions.

Earn the Allowed Rate of Return
A surprising number of utilities earn less
than their allowed rate of return for at
least a portion of their customer base.
The reason is that most utilities in this
situation look at their customers as one
uniform group, even though the failure to
achieve the allowed return may, in fact, be
due to low returns for certain customer
groups, geographic areas and/or products
and services. Customer profitability analysis
can reveal those segments with the
lowest returns so that the utility can take
a focused approach to addressing the root
causes and increase the likelihood of earning
the allowed rate of return.

Target the Marketing of New Products and Services
Most businesses selling goods and services
to the mass market perform extensive
market analysis to identify the targets
with the most potential. In contrast,
utilities whose offerings go beyond the
traditional electric and gas service often
possess limited information on which
they must base critical decisions. By
combining customer profitability analysis
with demographic data, utilities can
identify the most promising prospective
customers and more efficiently market
new products and services, resulting in
profitable revenue growth.

Focus Process Improvement and Cost Reduction Efforts
Utilities often use benchmarking to identify
functions with inefficient processes
and excessive costs. Unfortunately many
utilities have discovered an inherent
weakness in benchmark data: It can mask
financial losses driven by the unique
characteristics of each utility’s rate
structures, cost profiles and customer
behaviors. Customer profitability analysis
accounts for these profit drivers to better
pinpoint areas of inefficiency and opportunity.
Traditional benchmarking can also
miss some of the most costly inefficiencies
– those resulting in dissatisfaction
among the utility’s most profitable customers.
When combined with customer
satisfaction surveys, customer profitability
analysis can show which processes
are most troublesome to customers both
driving profits today and most likely to
buy new products in the future.

Taken together, these applications of
customer profitability analysis can provide
substantial value through reductions in
costs, increases in revenue and improvements
in customer satisfaction.

Profitability Analysis Approach

While there are several ways of talking
about customer profitability, we define it
in terms of Gross Margin, as follows:

Customer Gross Margin = Net Revenue
– Cost of Goods Sold – Transaction Costs

In this formula, Customer Gross Margin
is the estimated gross profit for each
customer. Note that this approach does
not take operating costs into account.
This is because a fundamental goal of the
analysis is to understand how customer
behaviors drive profitability and how the
utility can change the way it does business
to enhance customer profitability.
These customer behaviors and utility
actions will not typically drive operating
costs, but operating costs could certainly
be factored in to the calculation to measure
net profit.

Data flow for a customer profitability model leads to in-depth analysis.Net Revenue represents the amount
paid monthly by the customer for utility
services, as derived from actual billing
data, less any refunds or other items that
should be netted, and excluding taxes and
fees that are passed directly through to
the customer. Cost of Goods Sold measures
the cost of the electricity and/or
gas provided to the customer and can be
estimated at the individual customer level
from a combination of customer usage
data and electric and gas market prices.
Transaction Costs include meter-reading
costs, billing costs, demand-management
program payments and related costs,
and call center costs, all of which can be
allocated to each customer based on his
individual activities and attributes.

One key challenge in calculating a customer’s
gross profit is matching energy
usage with market prices to estimate the
cost of goods sold. Ideally, hourly meter
data for each customer’s usage would be
available, but this information actually
remains quite limited. Instead, a reasonable
estimate of energy usage can be
developed by analyzing representative
load curves by day of week and time of
day for various customer types and then
allocating total monthly usage data from
customer billing records into more discrete
time periods. If the representative
load curves are available on an hourly
basis, they can be used to estimate hourly
usage and then hourly prices for energy
can be applied to estimate cost of goods
sold. Alternatively the day can be divided
into periods of, for example, six hours, and
then the average energy price for each
period can be applied to estimate the cost
of goods sold.

The diagram in Figure 1 shows a highlevel
view of the typical data flow for a
customer profitability model. Customer
data is combined with unit cost data in
the analytics engine to calculate gross
margin, and then data on customer demographics
and satisfaction are merged to
populate the database storing the analysis
results and key supporting data. The
results and supporting data can then be
extracted for reporting and further analysis
outside of the model.

Typical Profitability Analysis Results

Customer profitability analysis results can be put into reports for further study.Once the customer profitability model is
developed and the data is merged and
analyzed, the results can be put into
reports to support further analysis. The
most basic report provides information
on the distribution of profitability for all
customers, as shown in Figure 2.

In Figure 2, derived from the actual
results of one electric and gas utility,
roughly 15 percent of customers prove
unprofitable, with the loss reaching up
to about $100 per month for some. This
is not too surprising, given that cross-customer subsidization is well-recognized.
The more interesting and actionable
results are revealed when the unprofitable
and the most profitable customers are
analyzed in detail.

Consider the diagram in Figure 3, which
compares the relationship between the
cost of goods sold and net revenue for
customers with the highest gross margin
versus those generating a loss. The result,
again based on a real electric and gas utility’s
data, demonstrates that transaction
costs are not a substantial factor in driving
profitability for these particular customers.
Instead, for the loss customers,
the high cost of goods sold is the primary
driver of negative gross margin. This suggests
that these customers may be particularly
heavy users of high-cost energy
during peak demand periods. Based on
this analysis, the utility should consider
targeting these customers for some kind
of demand-management program. In fact,
the utility can use the profitability model
to establish the level of demand-management
payments to these customers that
would result in an increase in profits.

This gas and electric utility discovered that the high cost of goods sold is the
primary driver of negative gross margin.In contrast, the high-margin customers
use a greater portion of their energy
during non-peak periods. This result
implies that demand-management payments
to these customers may actually
reduce profits. More detailed analysis of
the usage patterns of these customers is
required, but the customer profitability
model has already uncovered potentially
valuable insights into profit and loss drivers
as well as pointing to opportunities
for improvement.


Customer profitability modeling can provide
valuable insights into the factors that
impact profits and losses within a utility’s
customer base. These insights can, in
turn, drive specific actions to increase
profits by:

  • Changing customer behaviors to
    enhance profits through program and
    policy changes;
  • Focusing process improvements and
    technology investments where they will
    have the most bottom-line benefits; and
  • Targeting new products and services
    to those customers most likely to buy
    them under profitable terms.

Because most utilities already have much
of the required data, the costs of profitability
model development and analysis
are moderate. Therefore, we believe this
analysis easily pays for itself, and should
be considered a fundamental capability
for utility management and planning.