Welcome/What's New

Products and Services                     
   Utility Customer Databases
   Other Data Products
   Models and Forecasting
   Market Analysis
   Revenue & Cost Management
   Business Consulting

   Natural Gas
   Energy Technologies
   Energy Service (ESCO)

   Jackson Associates
   White Papers

Utility Service Area Database Issues

Electric Utility Service Area Databases provide critical market information on customers within each electric utility service area in the US. Service Area Databases include the same extensive firmographic and psychographic information on individual customers provided in the MAISY State-Level Databases (e.g., business type, number of employees, income, demographics, equipment and building structure detail, energy use including hourly load profiles) and can optionally include electric bills and measures of cost of service and profit for each customer record.

Electric bills reflect current utility rate structures and each customer's monthly and/or hourly electricity use; customer-detailed cost-of-service is determined from each customer's hourly electricity use profiles and both hourly generating costs and capacity cost allocations. Customer profit shows the difference between revenue derived from the customer through monthly bills and the cost of actually providing electricity to the customer. Use Service Area Databases to immediately calculate service area headroom, size and profile individual markets, assess revenue and profit potentials, analyze cost-of-service issues and evaluate competitive threats, opportunities and strategies. Databases are available for all electric utility service areas including investor-owned utilities, municipal utilities, public utility districts, and co-op areas. Electric Utility Service Area Databases use the standard MAISY chart-based drill-down software described above. The Service Area Profiler can also incorporate dynamic load profiling to determine monthly profit (or loss), revenue and cost of service for both residential and commercial customers and customer segments whose settlement is determined by distribution-utility dynamic load profiles and monthly meter readings.

While deregulation typically occurs throughout an entire state, existing rate structures, customer loyalty (or lack thereof), cost-of-service characteristics of the distribution utility, weather characteristics and other utility-specific factors distinguish each service area as its own market area. MAISY Service Area Databases are a cost-effective and timely source of information available for individual utility service area market evaluations and marketing analysis.

Use Service Area Databases to immediately:

Utilities and other Energy service providers (ESPs) preparing for future deregulation can use the Service Area Databases to develop effective marketing strategies before competition for customers actually begins while ESPs already operating in competitive markets can benchmark current marketing efforts and refine marketing strategies to improve profits.

See Texas Utility Marketing Databases Updates to see how MAISY products can be extended to meet state-specific competitive rules.


Customer information has little value in regulated markets; consequently, utility customer information sources are extremely limited. While traditional customer list sources such as Donnelley and Dun and Bradstreet (D&B) provide information on household income and demographics and on commercial and industrial firm employment and other variables, these data sources don't provide energy use information required to assess electricity sales profits. Complicating factors even more, establishment records in traditional list databases do not automatically correspond to utility customer data because of multi-establishment, single-metered buildings (e.g., each of the fifty companies in a medium-sized office building is an individual establishment in the list database; however, only the owner of the building is a utility customer ). Furthermore, customer counts provided by utilities (and reported in Department of Energy publications) reflect the number of meters and not the number of individual utility customers.

Customer information is critically important in competitive markets. MAISY Utility Service Area Databases are a comprehensive source of customer-detailed energy use, load profiles, firmographic and psychographic variables, revenue, cost-of-service and profitability measures.

MAISY Service Area Databases, which are available for all electric utility service areas (including municipal utilities, public utility districts, and co-ops), provide statistically representative samples of customers for each service area. Customer records are developed from MAISY State-Level databases and from additional data collected on individual customers within each service area. Electric bills reflect actual utility rate structures and each customer's monthly and/or hourly electricity use; customer-detailed cost-of-service is determined from each customer's hourly electricity use profiles, actual utility or market-based hourly generating costs and fixed capacity costs. Client-provided rate structures and cost-of-service data can also be incorporated in the databases. Customer profit shows the difference between revenue derived from the customer through monthly bills and the cost of actually providing electricity to the customer.

Electric Utility Service Area Databases use the standard MAISY chart-based drill-down software which answers specific marketing questions with mouse-clicks on charts and permits users to navigate through the state-level energy marketing and hourly load databases, exploring customer characteristics and relationships. Information can be viewed in chart or table form at any level of aggregation ranging from an individual customer to all customers in a state or states.

Users can also apply MAISY Profiler software which uses a "wizard approach" to assist in the analysis and evaluation of customers, customer segments and the utility market. This software is an add-on system designed to assist MAISY users with customer and market hourly load and profitability analysis. The profiler runs with State-Level and Electric Utility Service Area Databases and: (1) adds a "wizard approach" where a set of sequential screens help users define desired market segment or customer specifications, (2) automatically extracts and processes relevant customer data from the databases and (3) presents the results in a special Excel Workbook which integrates hourly loads, cost of service and rate structures. Use the Profiler for applications ranging from strategic market analysis to individual customer pricing and profitability evaluations.

Service Area Database Benefits

  • Provides immediate answers to critical utility market, market segment and customer questions

  • Provides the a cost-effective and timely source of utility customer information

  • Provides customer-detailed revenue based on current utility rate structures

  • Provides cost-of-service based on generation and capacity costs along with customer hourly loads

  • Supports bottom-line revenue, cost of service and profitability evaluations

  • Provides firmographic and psychographic information compatible with traditional marketing analysis and databases

  • Point-and-click software facilitates exploration and analysis

  • On-call telephone support

A Comparison of Traditional Utility Analysis and MAISY Utility Analysis

MAISY Utility Service Area Databases provide answers to a variety of difficult strategic marketing questions. This section illustrates the advantages of using detailed customer data in the MAISY Databases to conduct utility-level analysis compared to the traditional application of aggregate customer-segment load profiles. In this example we address essentially the same question from the perspective of two different competitive market players: the distribution utility facing competition and an outside ESP considering entering this utility market.

Questions for the distribution utility:

  • Which customer are high-profit customers and therefore most at risk to marketing and price offers from new ESPs?
  • Which customers will create a profit drag in a competitive market?

Questions for the ESP considering entry to the utility market:

  • Does enough "headroom" exist to support a marketing initiative in the utility service area?
  • Which customers are high-profit customers and should be targeted?
  • How much leeway exists among the target customers to support aggressive pricing?

To answer these questions Traditional utility service area analysis applies aggregate load profiles for broad customer segments, cost of service data and current utility rates to determine revenue and cost of service for each customer segment. If a dozen customer-segment load profiles are available for commercial sector analysis, for instance, results of revenue and cost of service computations for each of the dozen segments will identify segments with highest and lowest profits while the results across all twelve segments will determine whether sufficient headroom exists to justify entry by new ESPs. The results of this aggregate analysis are biased, however, because customer segments load profiles "cancel-out" all variability in individual customer revenue and cost of service that occur within the segment. This aggregation error can be quite large, depending on the intra-segment level of customer variability.

MAISY Utility Service Area Databases permit customer-detailed analysis avoiding the aggregation error inherent in the traditional analysis. The remainder of this section demonstrates the value of customer-detailed analysis in addressing the questions posed above.

The Utility Service Area - Rates and COS

To limit analysis details we focus on commercial sector customers for a utility whose service area encompasses thirty-five counties including two small metropolitan areas and a little more than one million total customers (about 2.5 million people). Of course, each utility service area is unique in its composition of customers, so quantitative results from this example analysis should not be extrapolated to other areas.

Monthly electric bills are determined with the following commercial rate structure applied to each commercial sector customer.

  • Less than 5kW maximum monthly demand: $11/month customer charge: $0.05/kWh for the first 1,000 kWh, $0.03/kWh for additional kWh
  • Less than 300kW maximum monthly demand: $35/month customer charge; demand charge of $4.00/kW, $0.018/kWh for all kWH
  • 300kW or greater maximum monthly demand: $350/month customer charge: demand charge of $7.35/kW, $0.0035/kWh for all kWH
  • With a fuel adjustment charge of $0.025kWh for all kWh

The average revenue across all commercial customers is $0.06/kWh.

MAISY Databases include cost of service (COS) developed individually for each customer based on the cost of supplying electricity in each hour of the day as determined from actual historical hourly generating cost and adding a capacity cost assessment based on the customer's monthly peak demands during each of five summer months. Alternative formulas can be used for capacity cost allocations; however, the results shown here are consistent with algorithms that assess capacity costs based on the contribution of the customer to capacity requirements. For simplicity in the analysis that follows, we use the distribution utility's COS to assess COS for the new ESP considering an entry to the market. This representation is reasonable if the new ESP contracts for power in the existing market; however, an ESP with its own portfolio of supply options will apply its own COS data. Other items can also be added to the COS data including metering, billing, marketing and other overhead expenses.

The Utility Service Area - Customers, Sales and Revenue

The figure below shows commercial customers in the utility service area by primary business activity. 58,763 commercial customers are located in the 35 counties of the service area (comparing similarly-sized service areas in DOE tables typically shows 50 -100% more "commercial customers" for a service area this size because the term "customer" in DOE tables is synonymous with meters, not actual utility customers. Many commercial customers have multiple meters and many of the meters reported as "customers " reflect minor non-building-related meters (parking lot lighting, etc.). The profile shown below is typical for a service area with a population of 2.5 million with two small metropolitan areas. Office and retail customers represent 44 percent (26,044) of all commercial customers in the service area.

The next chart shows the distribution of electricity use across primary business types revealing the more electricity intensive activities of offices, retail, malls, groceries, restaurants, hospitals and hotels.

Annual revenue, shown in the chart below, reflects annual kWh as well as demand charges especially for restaurants which tend to have large peak demands relative to annual kWh.

The charts above provide an overview of the service area; much more detailed assessments can be developed with MAISY by generating charts and tables of customer characteristics for any number of market segments within the primary business types.

For the commercial sector as a whole, we know that billing revenue from all customers is approximately the same as the cost of serving these customers since current electricity rates have been designed based on customer-class cost of service studies. This result is shown in the following chart where total commercial customer revenue for the distribution utility (annrev) of $493 million is nearly offset by cost-of-service (anncos) leaving a net profit (annprf) of only $3.5 million or $60/ per customer ( a "normal rate of return" is included in the cost of service data, so annprf reflects profit in addition to that allowed in the regulated rate of return).

Traditional Evaluation with Customer-Segment Load Profiles

Traditional analysis applies customer segment load profiles to address the questions posed at the beginning of this section. Sector-specific profit or headroom is shown in the chart below based on twelve aggregate customer segment load profiles developed from the representative sample of utility customers for the service area. The aggregate segment load profiles were applied to the same rate structures and cost of service data used in developing profit for individual customers

To answer questions posed at the beginning of this section, the distribution utility will designate the most profitable customer segments as "at risk" and the least profitable as its "problem segments." New market entrants will apply the same data, summing profit across all segments ($10 million) to determine whether potential revenue sufficiency exceeds costs to justify entering the market. If the new ESP decides to enter the market, it will focus its marketing efforts on customer segments with the greatest profit.

As we will show below, these aggregate segment results and the strategies that they suggest suffer from serious aggregation biases and lead to marketing strategies that are counterproductive; that is marketing strategies that tend to diminish commodity profits.

Customer-Detailed Analysis With MAISY Utility Service Area Databases

With the MAISY Service Area Databases we have access to detailed customer information for a statistically representative sample of customers in the service area. A quick look at revenue, cost of service and profit shows that the customer segment analysis of traditional market evaluations mask highly variable profit margins across individual customers. The table below shows revenue and profit just a few of the customers in this service area along with the customer's primary business category.

Since the Service Area Databases provide individual customer profits, we can develop accurate sector-specific profitability results directly from the customer data without having to apply aggregate customer-segment load profiles. We would not expect the aggregate load profiles to provide exactly the same results as computing profit from the individual customers because of aggregation bias inherent in the aggregate load profile process.* However, if intra-segment customer variability is small compared to variability across segments, the level of aggregation bias may be insignificant. The chart below presenting aggregate results along with the true individual customer calculations shows that aggregation bias is, in fact, quit significant. The true individual customer calculations were developed by computing profit for each of the 1076 customers in the representative sample and summing the profit from these individual customers within business types to get profits based on individual customer loads. Comparing these true profits to the biased profits based on customer segment averages reveals the inaccuracies of using the segment profiles shown in the second chart.

As these results indicate, profitability computations based on aggregate load shapes provide extremely poor estimates for all but one segment (assembly). For offices (off) and warehouses (dwr) sector-specific profit has the wrong sign. These results illustrate the inaccuracy of the segment-specific profitability estimates. Estimation errors for eight of the twelve segments are greater than 90 percent.

Summing profit for individual customers across sectors reveals a true total profit potential of $3.5 million. However, segment-wide results reflect the sum of profits, some positive and some negative, across customers within each sector. Evaluation of the individual customer profit data shows that the $3.5 million in profit for the entire commercial sector actually reflects $22 million in profit and $18.5 million in losses for a net profit of $3.5 million. The chart below shows profits for each customer quintile where customers have been sorted from least profitable to most profitable.

As indicated with these results, potential customer profits and headroom can not be reliably estimated with traditional aggregate customer segment analysis.

Answering the distribution utility and new ESP questions accurately requires analysis which includes revenue and cost of service at the customer level. It is not necessary to have information on all customers in the service area to conduct customer-detailed analysis; analysis of a representative sample of customers is sufficient to develop accurate service area revenue, cost of service, profitability and headroom assessments.

Customer Profitability Models and Individual Customer Profit Scores

The critical tasks for the distribution utility and the new ESP are to identify profitable and unprofitable customers and profit-based customer segments to gain the competitive advantage inherent in the questions raised at the beginning of this section. Customer profitability can easily be associated with customer characteristics such as number of employees and business type in order to match profitability analysis results with names of actual customers in customer list databases (e.g., D&B, Donnelley). These profitability relationships or models, which can be developed with cross-tab results or simple statistical models, provide a reliable guide to identifying profitable and unprofitable customers.

The chart below illustrates the ability of simple profitability models to distinguish customer profitability. In this case a profitability regression model was applied to MAISY utility customer data to determine how well the model actually distinguished between profitable and unprofitable customers. Actual profit is shown on the horizontal axis while predicted profit is shown on the vertical axis. The diagonal 45 degree line shows all of the points where actual profit equals predicted profit. Each dot on the chart shows a customer's actual profit and the profit predicted for that customer with the model. Those dots that lie closest to the diagonal line have been predicted most accurately. For instance, the right-most dot in the upper-right quadrant reveals an actual profit of $9,500 and a predicted profit of $10.000. All of the dots that lie in this first quadrant reflect customers for whom the model correctly predicted positive profits while the dots in the third quadrant (lower-left) reflect customers for whom the model correctly predicted negative profits.

Profitability relationships and models for other market segments are equally effective in distinguishing customer profitability. Applying profitability models to individual customer characteristics derived from commercially available customer list sources provides profit estimates or profit scores which can be used to identify the most profitable customers prior to their acquisition.

Thus, MAISY Utility Service Area Databases permit utilities and ESPs to more accurately answer the questions first posed in this section without any of the aggregation bias inherent in traditional aggregate load profiling approaches.

To recap:

  • High-profit customers can be identified from the MAISY Utility Service Area Databases which incorporate current utility rate structures and relevant cost-of-service data.
  • Customers who will create a profit drag in a competitive market can also be identified and targeted with ESCO, incentive rates or other measures to improve profitability.
  • Headroom can be more accurately evaluated using MAISY Utility Service Area Databases compared to the aggregate customer-segment load profiles.
  • The potential for competitive pricing can be determined from from the MAISY Utility Service Area Databases.

* Aggregation bias, in this instance, is a result of the fact that customer revenue and cost of service are both nonlinear functions of hourly loads. With nonlinear functions the value derived from an average input is different than the average of the values derived from the individual inputs. For example consider the function y = x2. With x values of 2, 4 and 10, we get an average x value of 5.33 which evaluates to 28.4 (i.e., 5.332). Evaluating each of the values individually provides three y values, 4, 16 and 100 whose average is 40.0. The aggregation bias in this example is 29% ( (40-28.4)/40 ). The true profit or headroom figure is determined by the profit provided by each customer; consequently, any discrepancy between the results based on aggregate load profiles and the results based on individual customer calculations reflects aggregation bias.

Using MAISY Service Area Databases to Identify Individual "High-Value" Customers

MAISY database records protect the identity of individual customers; however, more than three hundred customer characteristics are provided to map information from the databases on revenue, cost of service and profitability to customer contacts and addresses from commercially available lists such as Dun and Bradstreet, Donnelley, Claritas, Database America and others. Identify high-value target markets that match rates to cost of service with the MAISY databases and analysis software and then select a subset of the customer list data that most closely represent the most desirable market segments. This selective access of customer list information and marketing activities typically save three to four times as much as the cost of the MAISY Utility Service Area Databases.

Jackson Associates provides both telephone and inhouse support to MAISY clients on this important marketing program implementation process.

Utility Service Area Database Development Summary

MAISY Utility Service Area Databases are composed of a sample of utility customers whose energy use and other characteristics statistically represent the entire service area. Commercial sector databases (i.e. office , retail, restaurant, etc) typically include 1,500 - 2,500 individual customer records and residential sector databases include 800 - 1500 records. In addition to all of the customer information available in the State-Level Energy Marketing and Hourly Load Databases, Service Area Databases include electric bills, cost of service and profitability measures for each customer record.

Construction of the Utility Service Area Databases includes the following four steps:

Develop Information on All Commercial and Residential Customers Within a Utility Service Area. The first step in the Service Area Database determines the total number and composition of all commercial and residential utility customers within the service area. The number of commercial customers by business type, size category and age are determined from a variety of sources including commerce department databases, directories, associations and proprietary sources. Extensive information on residential dwelling unit type and age, income, demographics, heating fuel and other variables are derived from census data and updated with current population, housing information and other data.

Develop Service-Area Customer Samples. Data on current service area population characteristics developed in the preceding step are used to determine which customers to draw from the State-Level Databases in a way that the sample of customers statistically represents customers in the service area. Sample sizes of 1,500 - 2,500 for commercial and 800-1,500 for residential databases are developed to provide robust representational accuracy. In some cases, the number of state-databases customers located within service area boundaries is less than the target sample sizes; in this case, customers in nearby-areas are "borrowed" for the service area database. The energy use and hourly load characteristics of the borrowed customers are updated , if necessary, with statistical/engineering models to reflect any climate differences between the service area and the location of the "borrowed" customer.. This adjustment process was developed and refined in our utility database development projects conducted over the past 15 years. Convenience stores in Raleigh "look" just like convenience stores in Charlotte, after accounting for weather-sensitive end -use energy adjustments.

Compute Monthly Electric Bills, Cost-of-Service and Profit for Each Database Customer. Electric bills reflect current utility rate structures applied to each customer's monthly and/or hourly electricity use. Customer-detailed cost-of-service is determined from each customer's hourly electricity use profiles, actual utility or market-based hourly generating costs and fixed capacity costs. Client-provided rates and cost-of-service can also be applied during the database development process or after delivery of the Service Area Databases.

Customer Profit is calculated as the difference between revenue derived from the customer through monthly bills and the cost of actually providing electricity to the customer.

(c) 2007 Jerry Jackson. All rights reserved.