An Enlightened Approach to Budgeting and Goal Setting for the Energy Manager

John Avina, C. E. M.
Director, Abraxas Energy Consulting
June 13, 2012


Every year energy managers need to report how much energy they saved and whether or not they met their energy savings targets. Most energy managers present reports comparing their current year’s usage to energy usage from a previous year. Savings targets are typically a percentage of that previous year’s usage. Setting targets and comparing in this manner usually does not generate an accurate estimate of energy actually saved. Inconsistencies arise from year-to-year fluctuations in weather, occupancy, production or other factors, which can interfere with savings results. Instead, an energy manager’s performance should be determined by comparing current year usage to a normalized baseline, which represents how much energy the building would have used given current year weather conditions, production, occupancy, and base year usage patterns. Energy savings targets should be set based upon this dynamic baseline. Using dynamic targeting, variations in weather conditions, occupancy, production or other factors will not hinder the accurate measure of how much the energy manager saved, and whether energy savings goals were indeed met. This paper explains, with an example, the differences between using a static and dynamic (weather normalized) targeting to demonstrate energy savings.


Every year, the typical energy manager will generate, or receive from management, a fixed energy budget. Throughout the rest of the year, he will be held to this static standard. He may also be asked to save 5% from prior year’s energy usage, or may simply be provided with less money than the prior year. This method is simple for management to understand. Either the energy manager is meeting savings targets or he is not. Unfortunately, this method does not produce a realistic picture of how much energy the energy manager is saving.

Even though the building envelope, the building equipment, and the control set points themselves may not change from year to year in a static building, the energy usage may change drastically. This change may have nothing to do with the energy manager’s performance-there are simply other factors involved. A year with a hotter summer may require more air conditioning than a normal year. A year with a colder winter may require more heating than a normal year. Doubling the production in a manufacturing plant will increase the usage.

Energy Managers that allow themselves to be held to a static budget may succeed or fail as a result of factors that have nothing to do with their energy conservation efforts: weather, production or occupancy, for example. If even one of these factors is not in his favor, the energy manager may not meet his savings goal. How fair is that? There is nothing worse than knowing that you are doing a good job, but coming up short in the eyes of management because you are being judged by an arbitrary and inaccurate standard.

Budgeting VS. Targeting

It is understood that management needs to create budgets, and needs to make best predictions of future energy costs. No one can predict the future-but best guesses must still be made. The important point though, is that energy managers do not need to be held to these budgets when assessing and reporting on their job performance. The energy budget is a financial tool, not an energy management tool. Energy managers should set separate targets to which they can aspire. The Static Target, treated below, is just like a budget - a number that is fixed, and has no relation to the unmanageable factors which actually do affect energy usage, such as weather, occupancy, hours of operation and production.

An Example of the Failure of Static Targeting

As an example, consider Joe, a first year facility manager in Lincoln, Nebraska who tried to meet his 5% energy reduction target by shutting off his air handlers during unoccupied hours during the summer months. He started the strategy in July and continued through October. During the cold months, he did not shut off the equipment.

Imagine his surprise when viewing the bills and finding that there were no energy savings in the summer months. “How could that be?” Joe wondered. Figure 1 presents actual utility bills for this store. Notice, are no energy savings to be found.

Unfortunately for our energy manager, the summer of 2005 was a hot summer for the Upper Midwest. As a result, energy managers might have had a difficult time meeting their energy savings targets, as the higher temperatures led to increased air conditioning usage, overriding any real energy savings gained. As you can see in Figure 2, there was a large increase in cooling degree days[1] for Lincoln in 2005 as compared to 2004.

Figure 1: Comparison of 2004 and 2005 kWh usage and a Static Target. To meet the savings goal, 2005 usage must be less than the Static Target (which itself is 5% less than 2004 usage).

Figure 2: Actual Cooling Degree Days for Lincoln, Nebraska. Note the much hotter summer in 2005.

What Joe perhaps didn’t realize is that the hotter summer likely resulted in increased air conditioning usage, which obscured the savings he actually did achieve by shutting off the air handlers. But that doesn’t really matter, since he was being held to a Static Target. His manager took the prior year’s monthly usage, subtracted 5%, and held Joe to that standard. That sounded simple when Joe started the job. Unfortunately, due to the hot summer, Joe would have to save a lot more than 5% to offset the increased usage due to additional air conditioning.

This is only an example, but this is a real phenomenon that affects all energy managers whose buildings have space conditioning mechanical equipment. When using a static measure such as simple utility bill comparison, weather does affect how much savings you report. Hot summers and cold winters will eat into real savings. Conversely, cool summers and warm winters will exaggerate any true savings. For this reason, it doesn’t make sense to set energy usage targets based upon a prior year’s usage. A hot summer or a cold winter could preclude you from reaching your savings goal.

Since the bills already have the hot summer (and consequent additional air conditioning usage) already added into them, then it would be best if the target did as well. Then the affect of weather could be cancelled out from both the target and actual bills, leaving only savings. This is exactly what Dynamic Targeting does.


If Joe had used Dynamic Targeting, he would have seen savings as shown in Figure 3 instead. Shutting off the air handlers actually did save energy, and he was able to save 5.9% for the year. He (and his manager) just didn’t know it.

Figure 3: Baseline, Dynamic Target and 2005 usage. Target is 5% less than the Baseline usage. 2005 bills must be less than Dynamic Target usage to meet the 5% goal.

Static Targeting VS. Dynamic Targeting

It is best to allow your savings target to fluctuate with changes in weather-this involves weather normalizing your utility bills, and is called Dynamic Targeting. Using this method, you compare what your target usage should be (based upon current weather conditions) with how much the building actually did use.

There are a few steps involved in this process. Implementing the steps is not that difficult, but the concepts may be difficult to grasp at first because it is so different from simply comparing current year usage to a past year’s usage and applying a percentage.


Static Targeting
Dynamic Targeting
Changes in weather* … Can greatly affect reported energy savings and conformance to target Should have no affect on reported energy savings and conformance to target
Energy Target Is a fixed (static) amount, established before the current year starts Is a dynamic amount varying based upon weather conditions*
Target is based upon Base Year’s bill +/- a percentage An equation the takes into account weather* +/- a percentage
As a measure of Energy Savings … Is a poor representation of an energy manager’s performance. Can accurately represent an energy manager’s performance.

*variation could be due not just to weather, but to occupancy, production, or other factors that are found that significantly and consistently affect energy usage.

This extra work is worth the effort, because the dynamic method results in a more accurate portrayal of true energy savings, and can save the energy manager from reporting failure when, in fact, he succeeded. In some cases, not using dynamic targeting could cost an energy manager his job.

The steps for Static Targeting are:

  • Select a Base Year with which to compare future usage
  • Set savings targets as a percent of usage
  • Then, each month …

  • Compare current year usage to the target usage amount to see if you are meeting your goal.
  • Convert the actual, target and deviance amounts into Dollars.
  • The steps for using Dynamic Targeting are:

  • Select a Base Year with which to compare future usage
  • Set savings target percentage
  • Determine the relationship between weather data and utility usage during the Base Year
  • Then, each month …

  • Determine how much energy your facility would have used during the current year given current year weather conditions
  • Determine target usage amounts
  • Compare actual usage to the target usage amount to see if you are meeting your goal.
  • Convert the actual, target and deviance amounts into Dollars.
  • Dynamic Targeting can be done using spreadsheets, although utility bill analysis software greatly simplifies the task.

    The remainder of this paper will detail the procedure involved in using Dynamic Targeting.

    How to Perform Dynamic Targeting

    The steps for Dynamic Targeting as presented above are now detailed in this section:

    Step 1. Select a Base Year

    Whether you are using static or dynamic targeting, you still have to choose a Base Year to compare to. A common approach is to use the prior year. Sometimes, a specific year is chosen, such as 1990 was chosen as the Base Year for the Kyoto Protocol. If significant changes or retrofits were made to the facility and you wish to see the effects of those changes on your budget, you should choose a year prior those alterations. In any case, you will need a full year of utility bills (12 months) for your Base Year.

    Step 2. Set Savings Target Percentage

    Rather than have your supervisor give you a static list of usage or cost targets you need to beat, find out what percentage savings is expected of you. In our example, Joe, the energy manager, only had to save 5%. We call this percentage our Target Savings Percentage. The actual Savings Target can only be calculated each month after it happens, as the Savings Target varies with changes in weather (production, occupancy, etc.).

    Step 3. Determine the Relationship Between Weather and Usage

    This is the hard part. If your building has air conditioning equipment or heating equipment, then your building likely has a relationship to weather. In this step, all we are looking for is an equation that represents how your building uses energy. An example equation might be:

    Baseline kWh = (2064 kWh/Day * #Days ) + ( 72 kWh/CDD * #CDDs )

    The equation tells us that the building uses 2064 kWh for every day in the billing period, and for every Cooling Degree Day (CDD), the building uses 72 kWh.[2] In other words, short billing periods, will have fewer days, and thus less usage, while hot months will have more CDDs, and thus more usage. The Baseline Equation was generated using Base Year billing data and Base Year weather data, and represents how the building uses energy based upon Base Year usage patterns.

    The Baseline Equation can be simply generated using utility bill analysis software, and can be generated using spreadsheets as well (with a bit more effort). This normalization method is presented in more detail in papers and books cited in the References section of this paper.

    Step 4. Determine How Much your Facility Would Have Used

    Once you have the Baseline Equation, everything is simple from here.

    For each month in your reporting period, you will need to get a current-year bill, and determine the number of days and Cooling Degree Days (CDDs) associated with the billing period. If you are normalizing for other variables like production, schedule or occupancy, then collect this data as well.

    We know the Baseline Equation represents how the building uses energy based upon Base Year building usage patterns. How much would the building use given current weather conditions? To find this out, we plug into the Baseline Equation the current bill’s number of days and CDDs, and/or other variable data.[3] In this example, the billing period was 7/16/06 to 8/15/06, which is 31 days, 91,000 kWh usage, and 495 CDDs.

    Baseline kWh = (2064 kWh/Day * #Days ) + ( 72 kWh/CDD * #CDD )
    = (2064 kWh/Day * 31 Days) + ( 72 kWh/CDD * 495 CDDs)
    = 99,624 kWh

    Baseline kWh then represents what the building would have used based upon Base Year usage patterns, and Current Year weather conditions (and number of days).

    This is a Dynamic Baseline. If it were much cooler, (i.e., there were only 100 CDDs), the Baseline kWh would have been much less.

    Step 5. Determine Current Month Target

    In our example, we expected a 5% savings Target. So we take Baseline kWh and multiply it by 95% to get what our target usage amount was.

    Target kWh = Baseline kWh * (1 – Savings %) = 99,624 kWh * 95% = 94,643 kWh

    Step 6. Determine if you are Meeting Your Target

    First, let’s calculate energy savings.

    Savings = Baseline kWh – Actual kWh = 99,624 kWh – 91,000 kWh = 8,624 kWh

    Now for Targets-compare the current month bill to the current month target usage. If we are meeting the target, then Actual kWh will equal or be less than the Target kWh. Deviation is the difference between the two.

    Target kWh = 94,643 kWh

    Actual kWh = 91,000 kWh

    Deviation = Target kWh – Actual kWh = 3,643 kWh.

    For this month, our energy manager did indeed manage to meet his Target. However, if we just used a static target, then we would have seen,

    Target kWh = Base Year kWh *95% = 81,800 kWh * 95% = 77,710 kWh

    Actual kWh = 91,000 kWh

    Deviation = Target kWh – Actual kWh = -13,290 kWh.

    Using Static Targeting, the energy manager would not have met the Target at all, but would have been showing a large increase in usage. August 2005 was hot compared to August 2004 (Figure 2)-it follows that there was an increase in usage from August 2004 to August 2005.

    Step 7. Convert Energy Units into Dollars

    Since management is usually interested in Dollars, you will need to convert your energy numbers into Dollars. There are many ways to handle this. The simplest way is just to use blended rates.[4]

    From the current month’s bill, divide total cost by total usage, and apply this ratio to your savings.

    Current Bill kWh = 91,000

    Current Bill $ = $7,587

    Blended Rate = $ 7,587 / 91,000 kWh = $0.083 / kWh

    Now convert the energy numbers into Dollars.

    Savings $ = 8624 kWh * $0.083/kWh = $719

    Target $ = 94,643 kWh * $0.083/kWh = $7891

    Actual $ = $7587 (from bills)


    In order for energy managers to provide reliable metrics for setting and meeting energy savings targets, a dynamic energy usage target should be established. Static Targets are typically a percentage of a prior year’s usage. Variations in weather can drastically affect energy usage of HVAC equipment in current year bills, thereby rendering the comparison of current year bills to Static Targets useless. Dynamic Targets modulate in response to variations in weather (or other variables), thereby allowing the affect of fluctuations in weather (or other variables) to be removed from the energy savings equation. Dynamic Targets are established as a percentage of Baseline energy usage, which is determined using linear regression analysis. This process is described in the International Performance Measurement and Verification Protocol, and implemented in utility bill analysis software, and can be reproduced in spreadsheet applications.


    A Best Practice Guide to Measurement and Verification of Energy Savings, Australasian Energy Performance Contracting Association for the Innovation Access Program of AusIndustry in the Australian Department of Industry Tourism and Resources, 2004.

    ASHRAE (2002). Measurement of Energy and Demand Savings. ASHRAE Guideline 14-2002. Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

    Avina, J 2007, “An Energy Manager’s Introduction to Weather Normalization of Utility Bills”, Web Based Enterprise Energy and Facility Management Systems, by Barney Capehart, Fairmont Press, Chapter 33, pp.329-339.

    Avina, J, 2006, “Three Powerful Utility Bill Analysis Methods for the Energy Manager”, 2006 World Energy Engineering Congress Proceedings.

    2007 International Performance Measurement and Verification Protocol (IPMVP) Volume I on Concepts and Options for Determining Energy and Water Savings, Efficiency Valuation Organization

    Metrix Version 4.2 User’s Manual, 2007, Abraxas Energy Consulting, 839 Higuera St., Suite J, San Luis Obispo, CA 93401

    Sonderegger, R.C. 1998. “A Baseline Model for Utility Bill Analysis Using Both Weather and Non-Weather-Related Variables”. ASHRAE Transactions, Vol. 104, Part 2, pp. 859-870.

    About The Author

    John Avina, President of Abraxas Energy Consulting, has worked in energy analysis and utility bill tracking for over a decade. During his tenure at Thermal Energy Applications Research Center, Johnson Controls, SRC Systems, Silicon Energy and Abraxas Energy Consulting, Mr. Avina has managed the M&V for a large performance contractor, managed software development for energy analysis applications, created energy analysis software that is commercially for sale, taught over 200 energy management classes, created hundreds of building models and utility bill tracking databases, modeled hundreds of utility rates, and set up and maintained M&V projects for a handful of 500 to 1000 unit big box store chains. Mr. Avina has a MS in Mechanical Engineering from the University of Wisconsin-Madison. He is a member of the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE), the Association of Energy Engineers (AEE), the American Solar Energy Society (ASES), and a Certified Energy Manager (CEM).

    Abraxas Energy Consulting provides utility bill tracking and energy management services for its clients world-wide. In addition to providing a selection of utility bill tracking software packages for its clients, Abraxas Energy Consulting creates, maintains and analyzes utility bill tracking databases, trains its customers in energy analysis and software, and performs energy audits and measurement and verification for ESCOs and facility managers.

    [1] Cooling Degree Days are a simplified measure of how much cooling would likely be needed by a building for a period of time, such as a billing period. If you were calculating Cooling Degree Days for a billing period, then you would calculate Cooling Degree Days for each day and sum them up over the billing period. Cooling Degree Days are calculated for each day using the equation:

    CDD = [(Thi – Tlo)/2 – TBP ] x 1 Day

    where Thi = the daily high temperature, and Tlo is the daily low temperature, and TBP is the building’s balance point temperature. In the past 65°F was considered a balance point acceptable for all buildings, but now engineers recognize that every building is different, and have their own unique balance points.

    [2] Only cooling degree days were found to affect building energy usage in a consistent pattern and thus used as an independent variable in the Baseline Equation. Other potential variables, such as production, schedule, occupancy or others were not found to affect building energy usage in a consistent pattern, and thus not used. For most meters, it is usually the case, that only cooling degree days or heating degree days is used in the Baseline Equation.

    [3] Since only cooling degree days are used as an independent variable in the Baseline Equation, there is no need to gather any other variable data.

    [4] Although blended rates is the simplest way to assign costs, it is by no means the best. Explicitly modeling the rate, if you have time, is the best option.