Green marketing is dead! Long live “Better marketing”


Death of a Greensalesman

Eco-marketing, also a called green marketing, has been in our lexicon since the 60s. The time has come to put an end of this oxymoron. Green marketing doesn’t work for masses, and we have 60 years of data to prove it. The idea behind green marketing is a very noble one when you look at it from outside. But when you look at the data, the word “green/eco-friendly product” doesn’t make a customer buy a product.

First of all, the green ideology is against consumerism, the green movement is against making purchases in general. So the term Green marketing becomes an oxymoron. Second, only 12% of Americans are true green customers, 65% are lazy green and the rest don’t care. Green marketing is niche marketing, but how niche is it?

Green marketing is the hardest niche out there, true green consumers are very smart and they do everything in their power to not the buy anything. These two creates a huge gap between green marketers and regular marketer as far as resources go. When you don’t have money to market your products, your products are dead.

Another problem with green marketing is the history of it. In the past, Green marketers or people who pretended to be green has been using the word green or eco-friendly as a reason to sell “worst products” at a higher price. Older consumers think the word eco-friendly means it’s a worse product at a higher price.

Death of a Green Salesman

We have to stop marketing our green products as eco-friendly and move to “Better Marketing”. I can hear your say “But Tesla? but Elon Musk?”, They are not marketing their products as green. They are simply marketing their products as being better. Bigger companies have tried and failed at green marketing for years. Philips Lighting’s first shot at marketing a standalone (CFL) bulb was Earth Light, at $15 each versus 75 cents for incandescent bulbs. The product had difficulty climbing out of its deep green niche. The company re-launched the product as “Marathon,” underscoring its new “super long life” positioning and promise of saving $26 in energy costs over its five-year lifetime. Finally, with the U.S. EPA’s Energy Star label to add credibility as well as new sensitivity to rising utility costs and electricity shortages, sales climbed 12 percent in an otherwise flat market. Instead of green marketing they simply marketed their products as “better”.

In this article, we analyzed the effects of “better marketing” vs eco-marketing. We created a model to recalculate the value of better products. This paper presents a summary of research designed to provide sales and marketing representatives with a tool for communicating possible better product impacts(BPI) that may result from installed measures.

BPI Study Analysis

The value of energy efficiency investments extends beyond energy savings to include facility cost savings and production/revenue increases, non-energy impacts (BPIs). While rigorous quantification of better product impacts (BPIs) is becoming increasingly important for the evaluation of energy efficiency, BPIs can also provide valuable information for program implementers, looking to market program-supported technologies to end-users. This BPI marketing study demonstrates how BPI can help close the feedback loop between program evaluation and implementation. The current BPI Marketing study utilizes BPI information collected through a previous study, the 2012 Commercial and Industrial (C&I) Retrofit BPI Evaluation study which was the most comprehensive BPI evaluation studies to date. i, ii, iii To obtain estimates of BPIs associated with program sponsored measures, the 2012 C&I Retrofit BPI Evaluation study captured a robust set of data through in-depth interviews regarding the specific facility cost and revenue changes resulting from the installation of that provides insight to segment-specific marketing messages. The present BPI marketing study leveraged this data to provide program sales and marketing staff with valuable selling points to address the industry specific needs of customers and increase the value proposition of participating in efficiency programs. This study provided quantitative and qualitative evidence of the changes to internal and external labor, parts and supplies, waste management, sales revenue, and other reported cost and revenue changes that may result from installing energy efficient measures within various industries. It also captured important respondent information necessary to characterize BPIs realized by different types of customers.

What we analyzed

While BPIs have historically been used in program evaluation to provide a stream of benefits used in cost-benefit analysis of existing programs, BPIs also provide an excellent way to create the feedback loop from evaluation to implementation. This study a repurposed the data obtained through one of the most comprehensive BPI evaluation studies to date to provide valuable marketing information regarding non- energy benefits that may result from installing energy efficiency measures.

The overall goal of the BPI marketing study was to leverage data obtained from the Massachusetts 2012 C&I Retrofit BPI study to develop an industry-level analysis for program implementers, demonstrating the value proposition of energy efficiency programs beyond energy savings. Specifically, the re-analysis of the 2012 C&I Retrofit BPI study was used to:

  • Identify statistically significant BPI categories and sub-categories by industry and energy type (electric, gas)
  • Identify non-significant BPIs that are logically sound for any specific industry
  • Review and summarize quantitative and qualitative descriptions of BPIs
  • Present prominent BPI themes by industry for use in sales and marketing.

The BPI marketing study was designed to assist program implementers, who are engaged in the marketing of energy efficiency programs, to understand the specific cost and revenue changes reported by customers within 13 separate industries. This study helps close the feedback loop between program evaluation and implementation by leveraging data used to quantify BPIs for the evaluation of programs and then re-analyzing that information to provide program sales and marketing staff valuable selling points that appeal to specific customer segments and increases the value proposition of participating in efficiency programs. This information will benefit utility programs that are implementing targeted marketing campaigns and niche programs to reach energy saving goals. We provide quantitative and qualitative information of the changes to internal and external labor, parts and supplies, waste management, sales revenue and other reported costs, and revenue changes that result from the installation of energy efficient measures within each industry.

Review of 2012 C&I Retrofit BPI study

In 2012, the Evaluation team conducted a BPI study that successfully captured data from more than 500 C&I program participants regarding BPIs resulting from 788 prescriptive and custom electric and gas measures. While the purpose of the present BPI Marketing study was not to provide an extensive review of the data collection process used in the 2012 C&I Retrofit BPI Evaluation study, the following summary of the data collection process offers the necessary context to understand the approach used in the study. Our approach to estimating BPIs broke impacts into mutually exclusive categories that reflect separate cost and revenue (business impacts) resulting from the Program Administrators’ (PAs) program-sponsored installed measures. Rather than rely on stated BPI estimates alone, we probed respondents for a deeper understanding of the impacts on specific costs and revenues. The key components of the methodology were as follows:

For prescriptive measures, we selected the sample of 297 and 153 electric and gas target completes, respectively, from the pool of respondents to the 2010 Massachusetts C&I Free-ridership/spillover study. Of those targeted completes, we were successful in completing 302 and 99 BPI interviews. For custom measures, due to the limited sample of custom projects in the 2010 Massachusetts C&I Free- ridership/spillover study, we supplemented the pool of respondents to the 2010 Massachusetts C&I Free- ridership/spillover study with an additional sample from the population of custom measures installed in 2010. The target number of completed interviews for custom measures was 276 and 112 custom electric and gas completes of which we were successful in completing 310 and 151 BPI interviews.

  • Designed the research instruments, trained the interview staff, and oversaw quality control;
  • Experienced DNV GL energy analysts conducted the semi-structured interviews;
  • Collected data on BPI types and dollar values, and like and unlike spillover.
  • Used ratio estimation to calculated BPIs by reporting measure category or end-use, as was needed by the Massachusetts Program Administrators for their cost-benefit analysis.

The evaluation team captured BPI information for 789 prescriptive and custom electric and gas measures. Positive BPIs or non-energy benefits were realized for 58% of measures, while 3% of measures resulted in negative BPIs. An additional 40% of measures reported no positive or negative BPIs.

The BPI question battery was structured to prevent possible double counting across 13 BPI categories by presenting related categories sequentially for easier respondent recall. The evaluation team used additional closed-ended questions to assess whether the respondent experienced an increase or decrease in each affected BPI (e.g., an increase or decrease in operations and maintenance costs). Next, we used open- ended questions to ask respondents to provide the overall dollar impact associated with each BPI category. In some instances, respondents were able to provide BPI values and articulate reasons for these values without further probing. However, in most cases, interviewers needed to guide respondents through a series of structured probes to determine whether respondents experienced any changes to various cost or revenue centers associated with each BPI category. For example, internal labor and external labor are separate cost centers associated with Operation and Maintenance (O&M) costs. Once the interviewer identified the impacted cost and revenue centers, deeper probes were used to determine the nature of those changes and specific metrics for quantifying the monetary and resource impacts of the installed measures. Table 1 presents the general probes for each BPI section.

Table 1. Non-Energy Impact Categories


Examples of specific probes for each BPI category included the following:

  • Operations and maintenance: The interview guide included probes for internal labor, external labor, parts, training, fuel saved, and other O&M costs. For the labor and training subcategories, the interviewers attempted to get annual hours of increase or decrease and an hourly rate. For parts, interviewers attempted to quantify the number and type of parts that increased or decreased and the unit cost of each. For training, interviewers attempted to quantify increases or decreases in training costs and whether these were one-time costs or recurring costs. For fuel, interviewers attempted to quantify specific changes (increase or decrease) in fuel usage. For other, interviewers asked the respondent if there were any other O&M related costs that increased or decreased that we had not yet covered.
  • Administrative or other labor: This section included probes for internal labor, external labor, training, and other. The use of these specific probes was similar to their descriptions in the O&M section, except they were applied to administrative rather than O&M costs.
  • Cost of supplies, materials and materials handling: This section included probes for internal labor, external labor, and other. The use of these probes was similar to their descriptions in the O&M section, except they were applied to materials handling rather than O&M costs.
  •  Waste disposal costs: The section included probes for waste materials, waste handling, permits, and other. For waste materials, the probes attempted to identify the type of material (e.g. carbon dioxide, sulfur dioxide, etc.) and quantify in units the increase or decrease in emission. For waste handling, the probes attempted to identify number of hours of labor, and the fully-loaded hourly costs for that labor, just like the internal or external labor costs in the O&M section. In the event that the respondent could not supply fully loaded hourly costs, interviewers attempted to gather enough information to allow us to impute it.

We used information collected through these interviews to develop standard formulas and metrics for each cost and revenue center (i.e., the cost or revenue items) impacted under each BPI category. Standardizing the formulas across multiple measures allowed analysts to evaluate each in terms of the necessary metrics (e.g., salary, hours, price), and the range of responses to those metrics ($/hour). For example, Table 2 shows the standardized formulas for the Operation and Maintenance costs category, the most widely referenced BPI category.

Table 2. Example of Formulas Used to Calculate BPIs: Operations and Maintenance Cost Changes Measures BPI Category Cost/Revenue Center Formula


We used these data to construct a robust database of BPIs that we then used to estimate statistically significant BPI estimates for 15 of 22 measure categories or end uses reported in the C&I BPI evaluation study. The respondent’s knowledge of BPIs suggests that BPIs can be effective tools in marketing participation in energy efficiency programs. Positive BPIs result from decreased costs for maintenance, administration, and waste management. Similarly, positive BPIs also result from increased revenues from added sales or production increases. In constructing the database needed to produce BPI estimate for evaluation, 2012 C&I Retrofit BPI Evaluation Study produced a wealth of information that could assist program implementers and utilities to help promote energy efficiency programs and target customers. Rather than allow this information go to waste, the Evaluation team coordinated with program implementation to mine the database for insights that assist in marketing efforts. This was the basis for conducting the BPI Marketing study, which is discussed in the remainder of this paper.

Marketing Methodology

In this section, we provide an overview of the steps in developing the marketing analysis. The marketing analysis relied exclusively on the repurposing of data collected through the 2012 C&I retrofit BPI evaluation study. Prior to developing the marketing analysis, the Evaluation team solicited input from the PAs implementation staff to ensure the analysis provided implementation staff with the most valuable information for their marketing efforts. Through this effort we determined that the marketing BPI analysis should reflect the following criteria for reporting results:

  1. BPI results should be presented by industry segment rather than measure category or end use – Implementation staff stressed the importance of industry-specific over measure messaging. To provide analysis at the industry level, the BPI marketing study leveraged self-reported industry classification from the obtained through the 2012 C&I retrofit interview process. Table 3 presents the number of measures used in the analysis by industry segment.

Table 3. Number of Customers Used to Estimate BPIs by Industry Segment*


* Note: Customers providing responses for prescriptive and custom, electric and gas measures are combined


  1. The unit of analysis should reflect distinct customers rather than individual measures – The unit of analysis for the 2012 study was the measure level because the PAs were interested in estimates of BPI per unit of energy savings for various measure categories or end uses. This worked well for BPI estimates used in the PA’s cost-effectiveness models that incorporate BPIs at that at level. However, for marketing purposes, it is more appropriate to report BPIs at the customer level than the measure. From the end user’s perspective, customers within an industry are likely to make similar energy efficiency investments, so the average cost and revenue changes resulting from that bundle of efficiency measures is more likely to represent the expected BPIs across the average bundle of measures. From the implementer’s perspective, rather than focus on a predetermined measure-level solution, implementation staff typically approaches customers with the objective of saving energy for the customer. This often consists of a multi- measure, custom, or comprehensive design approach. Therefore, the team determined that collapsing BPIs from the measure-level to the customer level was more beneficial for marketing purposes. However, using custom-level rather than measure-level results decreases the sample size and deflates the standard error from the customer’s perspective. In making this change, the Evaluation team needed to re-assess the statistical significance of BPI estimates at the customer level. As seen in Table 4, changing from the measure to the customer level substantially reduced the number of observations in the analysis used to compute BPIs, potentially reducing statistical significance of BPI estimates.

Table 4. Number of measures and customers with non-zero BPIs by BPI sub-category 6


  1. Presented BPIs in terms of the BPI sub-categories rather than BPI categories – The 2012 C&I retrofit BPI study showed that subcategories (internal labor, external labor, parts, and supplies, etc.) were more recognizable to end-users than the BPI categories. While the latter were beneficial to ensure that respondents did not double count cost and revenue changes realized in different business units, respondents frequently conceptualized the BPIs in terms of the sub-categories. Table 4 below shows the relationship between BPI categories and sub-categories. DNV GL re-organized the BPI estimates to provide estimates at the sub-category level that more closely reflect meaningful facility changes for marketing purposes. While other BPI sub-categories were also captured by the interviews, the frequency in which these categories were reported was very limited. Therefore, our analysis focused on internal and external labor, parts and supplies, and training cost changes.

Table 5. Mapping BPI Sub-categories to BPI categories



  1. Identify statistically significant or logically consistent BPIs by industry and BPI sub-category We aggregated the BPI estimates to the industry, customer, and sub-category level then used the statistical procedure of ratio estimation to develop estimates of BPI per kWh or per therm, for electric and gas measures, respectively. We then expanded the sample results to the population of measures. This was accomplished by calculating the ratio of BPI (in dollars) to reported savings for the sample for non-zero BPI values. The calculation of the BPI adjustment factor used appropriate weights corresponding to the sampling rate. The adjustment factor was calculated as a ratio estimator over the sample of interest (Cochran, 1977, p.165). The formulas for these factors are given below. The BPI rate RI was calculated using:



GTj = tracking estimate of gross savings for measure j

GIj = evaluation estimate of gross nonenergy impacts for measure j

wAj = weighting factor for measure j used to expand the sample to the full population

After calculating the average BPI per kWh and BPI per them and their respective standard errors, we identified BPI sub-categories within each industry segment that were statistically significant at the 90% confidence level. For those BPI subcategories that showed non-zero BPIs that were not statistically significant, we reviewed the interview responses to determine whether the qualitative descriptions provided logical explanations for changes a facility in a particular industry might expect.

  1. Summarize BPI results by industry and BPI sub-category – Once we identified the statistically significant BPI sub-categories within each industry, the next step was to summarize the survey data for those sub-categories. For the BPI marketing analysis, we summarized both quantitative and qualitative BPI results. These are discussed below.
  • Quantitative BPI results –We calculated the average BPI per kWh savings for electric measures and average BPI per therm savings per customer for gas measures at the BPI sub-category level for each industry. In order to show end-users the magnitude of the BPIs, we needed to impute a value for the energy savings. Therefore, we estimated high-, medium-, and low-point estimates for expected BPIs by assuming the 75th, 50th, and 25th percentile of energy savings.
  • Qualitative BPI results – To assist sales and marketing personnel in “telling the story behind the numbers,” DNV GL provided anecdotal information to support the quantified BPI estimates. During the interview process, energy analysts captured detailed information concerning the nature of changes to business operations resulting from the installed measures.


Figure 1 and Figure 2 below present the median BPI for electric and gas measures, respectively by industry. The figures also show the percent of each industry’s BPIs that are derived from internal labor, external labor, parts and supplies, and training. From Figure 1, we see that the highest median BPIs are reported by the public order/public safety, manufacturing and office segments. We also see that the sources for those BPIs vary considerably across these three segments, parts and supplies make up the largest share of the average BPI in the manufacturing sector, while internal labor constitutes nearly 1⁄2 of the BPIs from the other two sectors. External labor constitutes the largest share of the BPI from the public order as a greater share of maintenance services for public institutions are likely outsourced compared to in the private sector. Further, facility staff in manufacturing settings are more often skilled in repair and maintenance of mechanical systems. Figure 2 shows that BPIs associated with gas measure BPIs range between $0.40 per therm and $1.00 per therm. External labor cost savings are the primary source of BPIs within the Education, Lodging, and Mercantile sectors. Figure 2 shows that statistically BPIs associated with gas measures were limited to just two industries, religious worship, and lodging.

Figure 1. Electric overview of industries with statistically significant BPIs*

* Median BPI presented for each industry = BPI/therm savings X Median Savings for the industry

Figure 2. Gas Overview of Industries with Statistically Significant BPIs

* Median BPI presented for each industry = BPI/therm savings X Median Savings for the industry

The following summarizes our analysis of the qualitative information captured by the 2012 C&I Retrofit BPI study provides the following insights into the specific cost and revenue changes within a number of industry sectors. We selected industries that were of particular interest to the PA’s implantation staff.

Sources of BPIs in Education

The most prevalent BPIs in the Education sector resulted from following changes:

  • Parts and supplies –New fixture required fewer bulb purchases; new HVAC requires fewer spare parts, saving on inventory and storage space; new windows led to no replacement of hardware on windows; building shell improvements eliminated the need to purchase new insulation on a yearly basis.
  • Internal labor – New lighting required less frequent replacement due to longer useful lives.
  • External labor – Occupancy sensors save from cleaners not having to turn off the lights; new HVAC measures decrease the need for specialists to fix the system; new equipment means few repairs.


Sources of BPIs in Healthcare

The most prevalent BPIs in the Healthcare sector resulted from following changes:

  • Parts and supplies – No more need to replace thermostats due to electronic sensors; fewer lighting components needed on hand.
  • Internal labor – Staff spends less time to coordinate with vendors; less time spent changing out lighting ballasts and fixtures; fewer bills to pay which saves on invoicing time; less time receiving and stocking equipment.
  • External labor – New system requires fewer contractor visits due to web monitoring.

Sources of BPIs in Lodging

The most prevalent BPIs in the Lodging sector resulted from following changes

  • Parts and supplies – New machines and systems require less day-to-day maintenance; Technological improvements mean that issues can be solved remotely; new products last longer than old versions; Motor belt lives increase from 6 months to 18 months.
  • Internal labor – Building shell improvements resulted in decrease time maintaining building and less ice damage and leaks to deal with during the winter; new HVAC system requires less management and repairs; HVAC controls eliminate the need for staff to set temperatures manually; problems can be fixed remotely and there is no longer a need to send someone to the problem area.; newer lighting last longer and do not need to be replaced as regularly.
  • Administration – Fewer issues require less paperwork to complete.

Sources of BPIs in Food Sales

The most prevalent BPIs in the Food Sales sector resulted from following changes

  • Parts and supplies –Less product spoilage/product loss.
  • Internal labor – Decrease restocking time, saved over 50% in costs, lower costs associated with stocking shelves due to less spoilage, less general maintenance (Electric).
  • External labor – Decrease number of contractor visits.

Sources of BPIs in Manufacturing

The most prevalent BPIs in the Manufacturing sector resulted from following changes:

  • Internal labor – Electrician used to have to vacate 150 people for 30 minutes 6 times a month to change out fixtures and ballasts(electric); no longer running around checking on equipment due to complaints; less time spent maintaining lighting; fewer internal repairs with high-end equipment; overall increase in worker activity due to new lighting; less lost down time due to machines breaking down.
  • Administration – Installed measures resulted in less time processing work orders.


The most prevalent BPIs in the Office sector resulted from following changes:

  • Parts and supplies – Have not had to purchase any replacement bulbs since new installation.
  • Internal labor – New lighting has fewer fixtures to maintain and requires less time spent changing out bulbs, this also results in less time filling out paperwork and processing complaints; controls do not require maintenance of thermostats.
  • External labor – New HVAC system is more reliable which saves on maintenance.
  • Revenue – Increased comfort due to more consistent heating; decreased turnover of the building and fewer tenant issues.


This paper presents a summary of research designed to provide sales and marketing representatives with a tool for communicating possible BPIs that may result from installed measures. Our analysis repurposed a robust dataset developed for the evaluation of energy-efficiency programs to provide program implementers with industry-specific BPIs presented in terms that end users actually think about their business needs. For many of the BPIs identified, the primary sources of value were derived from internal and external labor cost savings or reduced need for parts and supplies. These cost savings largely result from the relatively high quality of energy efficient equipment. As programs incentivize participants to use energy efficient appliances, fixtures, or controls, they are installing high-end equipment. That equipment tends to last longer and requires less maintenance. This translates into direct cost savings for internal and external labor, but also lower costs for parts and supplies, and reduces administrative costs for ordering and processing invoices. This analysis clearly demonstrates that the value proposition for program implementers and even equipment manufacturers extends far beyond energy savings. Finally, this analysis offers an example of a successful feedback loop between program evaluation and implementation.


We recognize the following limitations of this study that should be considered when leveraging material contained in the presentation:

  • Values and anecdotal information presented in this study reflect the mix of measures and participants contained in the 2011 program tracking data. The “types of changes reported” should serve as talking points of what other participants have experienced, given the measures they installed and the specific operating procedures at their facility.
  • Due to substantial variation in the magnitude of gas BPIs and the relatively limited sample size, this study was not able to identify statistically significant BPIs for gas measures for most industry sectors. However, the 2012 C&I retrofit BPI study did find statistically significant BPIs for a range of prescriptive and custom measure categories. Further analysis would be required to combine BPIs across similar industry segments to provide statistically significant results the at the most appropriate level of industry segmentation.
  • The following factors may limit the applicability of BPI estimates in other jurisdictions:

o Values were specific to Massachusetts customers. For example, the general cost of labor in Massachusetts may be higher than that in a Midwestern state.

o The mix of measures assumes C&I programs that are retrofits, which consisted of a mix of early replacement and replace on failure measures. Additional steps should be taken to address new construction.

  • Certain territories in Massachusetts that are demographically unique should consider adjusting BPI estimates to reflect differences in the mix of measures within the territories relative to the state as a whole.


Final Report – Commercial and Industrial Non-Energy Impacts Study. Prepared for the Massachusetts PAs. Prepared by DNV GL and TETRA TECH. June 29, 2012. ii

Stevens, Noel, Lindsay Foley, Susan Weber, Pam Rathbun, and Miriam Goldberg. Using In-depth Interviews to Estimate Non-energy Impacts Resulting from Commercial and Industrial Energy Efficiency Measures. Paper presented at IEPEC. Chicago. 2013. iii

Stevens, Noel, Kimberly Crossman, Pam Rathbun, Miriam Goldberg, and Ben Jones. Non-energy impacts of C&I energy efficiency measures provide substantial program benefits above carbon reduction. Paper presented at IEPPEC. Berlin. 2014.


Leave a Reply