�ݺ�ߣ

THE REVENUE NEUTRAL SALES MODEL: A NEW APPROACH
TO ESTIMATING LIGHTING PROGRAM FREE RIDERSHIP
Presented at the International Energy Program Evaluation Conference –
Chicago 2013
Tami Buhr, Opinion Dynamics
Stan Mertz, Applied Proactive Technologies

Overview
IEPEC Chicago 2013 2
 If a lighting evaluator were
granted three wishes, what
would he wish for?
 Why existing methods for
estimating lighting program
free ridership are problematic
 Describe a new model that
grants the evaluator’s wish
and the theory behind it
 Example of the model in
practice
IEPEC Chicago 2013

The Challenge of Evaluating Upstream Lighting
Programs
 “Participants” disappear
when they make purchase
and walk out of retailer
 Program data contains
number of bulbs sold but
not who bought them
 Cannot contact
participants at end of
program year and conduct
NTG self-report surveys

Three Wishes
 Evaluators wish they had sales
data to calculate lift in sales due
to discount
 Pre and post-program sales data
 Comparison area or store sales
data
 Complete lighting category sales
data
IEPEC Chicago 2013

Evaluators Have Not Been Granted Their Wish
 Retailers will only provide
sales of program-discounted
bulbs
 Sales data is a trade secret.
Provides clues to a retailer’s
merchandising strategy that
could be used by
competition.

What Have Evaluators Done?
 Attempted to make traditional
methods work with mixed
results
 Wide ranging NTG results
 Large confidence intervals
 Results are contentious
 No one is happy

Attempts at Self-Report
 General population telephone surveys
 Call utility customers and ask detailed questions about past lighting
purchases
 Results are of questionable validity due to timing of survey, small
nature of purchase, and difficulty identifying program purchasers
 In-store customer interviews
 Interview customers in store immediately after they make purchase
decision
 Greater confidence in self-report results
 Challenging to get retailer permission to conduct
 Usually make use of non-probability samples
 Survey mode may heighten social desirability bias

Other Methods Attempted
 Retailer Interviews
 Becomes another request for sales data that is denied
 Store level staff often do not know sales
 Corporate level do not know for a specific utility territory
 At best, get rough estimates
 May have vested interest in seeing programs continue
 Modeling Techniques
 Many require use of self-report data in addition to other data
(e.g. multi-state model, revealed preference models)

There’s Hope!
Using the Sales Data We Have
 We have program sales data
 From the contracts retailers sign
with the utility (MOUs), we also
have:
 Program discounted price
 Full price
 Number of bulbs utility will
discount
 Based on an understanding of
retailer decision making, we use
the data we have to estimate
what sales would be at full price

Retail Accounting and Upstream Programs
 Total Revenue = Price * Quantity Sold
 Total Profit = Revenue - Costs
 Retailer income statement
 Revenue on Top Line = “Topline Sales”
 Profits on Bottom Line = Company’s “Bottom line”
 Retailers can only claim the actual sales price as revenue on their
topline
 Reimbursement from the utility for the discount cannot be claimed
as revenue.
 Reimbursement goes into profits a reduction in cost of goods sold
 Profit = Revenue – Costs + Utility Reimbursement

What Happens to Retailer Profits and Revenue when
Participating in an Upstream Program?
 Profits?
 No problem.
 Retailers are reimbursed for discounts and added
to profits.
 If retailer sells one additional bulb, profits increase
 Revenue?
 We might have a problem.
 Can only claim discounted sales price as revenue
 For each unit sold with the program, less revenue
than before the program
 Retailer revenue could drop if sales do not increase
enough to cover lost revenue due to the discount

Revenue with and without Upstream Program
$-
$100
$200
$300
$400
$500
$600
$700
$800
100 125 150 175 200 225
Revenue
Quantity
Revenue With Program ($3 per bulb)
Revenue With Program ($2 per bulb)
Revenue Neutral Point.
Without the program discount, 100
bulbs sold at $4/unit for revenue of
$400.

Profits and Revenue Matter to Retailers
 Upstream programs are good for a retailer’s profits
 Retailers still care about revenue
 Wall St. looks at change in revenue (e.g. year over year growth,
same store sales comparisons)
 Buyers’ and managers’ bonuses are based on revenue
 Unhappy if utility program causes revenue to drop
 Retailers structure their MOUs with utilities so at minimum
the revenues remain the same.
 Program impact must be revenue neutral
 Theory confirmed by interviews with biggest nationwide
corporate retailers

Retailer Decision Tree

Model Implementation
 If retailers structure their MOUs to be revenue neutral with the program,
can use a reverse calculation and the data we have to estimate their
revenue without the program
Price (P) = R/Q Allocation/Sales
(Q) = R/P
Revenue (R) =
P*Q
Without Program $5.85 ?? ??
With Program $1.85 4,000 $7,400
Price (P) = R/Q Allocation/Sales
(Q) = R/P
Revenue (R) =
P*Q
Without Program $5.85 1,265 $7,400
With Program $1.85 4,000 $7,400
Data from MOU
Planning Results

Planning Free Ridership
 Free Ridership = Units without Program/Units with Program
FR = 1,265/4,000
FR = 0.32
$-
$1,000
$2,000
$3,000
$4,000
$5,000
$6,000
$7,000
1265 units @$5.85 4000 units @ $1.85
Revenue
Units Sold at Regular and Program Pricing
Need to
sell 2,735
more units
at $1.85
for
revenues
to remain
neutral
1,265
units would
have been
sold at
$5.85

End of Year Free Ridership
 Fell short of goal for product. Sold 2,535 bulbs.
FR = 1,265/3,800
FR = 0.33
$-
$1,000
$2,000
$3,000
$4,000
$5,000
$6,000
$7,000
1265 units @$5.85 3800 units @ $1.85
Revenue
Units Sold at Regular and Program Pricing
Sold an
additional
2,535 units
due to price
drop to
$1.85
1,265 units
would have
been sold at
$5.85

Model in Practice
0
5000
10000
15000
20000
25000
30000
Bulbs
Removeddiscounts on
top‐selling spirals
Reinstated discounts but at lower
rate on some products
Delaware Department of Natural Resources and Environmental Control
Lighting Program
Period 1 FR = 0.39
Period 2 FR = 0.60
Period 3 FR = 0.49
Overall Program FR = 0.51

Model Advantages
 Based on all sales from all retailers
 Can estimate free ridership by:
 Retailer
 Bulb Type
 Time Period
 Can be used as a planning tool
 Have an estimate of free ridership before program year
 Can make midyear corrections if not meeting goals
 Information can be used to improve program
 Analysis is inexpensive to conduct relative to other methods

Model Challenges
 Data tracking is important
 For each product sold, need:
 Regular pricing
 Program pricing
 Program allocation
 Number of units sold
 Price changes
 Is an estimate of the maximum free ridership
 Actual free ridership could be lower
 Currently only considers influence of program pricing and not
other program efforts such as education or product placement

Conclusion
 Upstream lighting programs are a key component of many
utilities’ residential program portfolios
 All methods for estimating lighting program net-to-gross have
strengths and weaknesses.
 The Revenue Neutral Sales Model is based on a verified
theory of retailer behavior that allows us to estimate what
sales would be without the program
 The method has a number of advantages that other methods
do not

Questions and Comments?
Tami Buhr
Director of Survey Research
Opinion Dynamics
tbuhr@opiniondynamics.com
617-301-4654
Stan Mertz
Director of Retail Operations
Applied Proactive Technologies
stanm@appliedproactive.com
413-731-6546

�ݺ�ߣ

IEPEC_The Revenue Neutral Sales Model_Buhr

Convert to study materialsBETA

More Related Content

IEPEC_The Revenue Neutral Sales Model_Buhr