Marketing Optimization

By Jerry W. Thomas

Marketing is a tricky business and

a dangerous career. It’s almost
impossible to measure the effects of
advertising, packaging, distribution
channels, media expenditures,
social media Likes and Tweets, sales
organizational structure, etc., on
brand share or sales revenue.

Without good data and absent any trustworthy feedback loop, marketing managers often turn to As a company
the security of marketing myths, pop culture marketing fads, fawning at the feet of consultants, and learns what
polishing up their résumés before the ax falls. drives its
Is marketing solely a game of chance, or might there be a way to bring scientific methods to the individual brands,
table? Let’s draw a distinction between the micro and the macro. At the micro level, the various commonalities are
pieces of the marketing puzzle can be, and should be, optimized on an on-going basis. The overall often found that
positioning and strategy should be evaluated. Every ad and commercial should be tested for make it easier,
effectiveness. Products should be tested and optimized. Promotions should be tested. Package and less costly, to
designs should be tested. Brand names should be evaluated. build marketing
models for its
These micro-level tests must be a constant and ongoing process of evaluation, tweaking, and
other brands.
reevaluation, to continuously improve the gears, bearings, and levers that make up the marketing
engine. Optimizing these micro elements of marketing typically yields improvements in sales
revenue and market share. But this is only the first step on the optimization stairway.
What happens when all of these elements are put together? How should the budget be allocated
among the different marketing elements? How should the budget be allocated geographically? By
different media? What is the optimal pricing strategy? What’s the optimal level and timing of media
advertising? How much money should be spent on extra salespeople versus increasing media
advertising? These are the macro types of questions. These questions cannot be answered by
copy testing, product testing, or other micro-testing methods. The workhorse of macro optimization
is marketing mix modeling.

1.817.640.6166 or 1.800. ANALYSIS • www.decisionanalyst.com

Copyright © 2016 Decision Analyst. All rights reserved.

Marketing Mix Modeling absolutely essential to successful modeling. The data must
What exactly is marketing mix modeling? The term is be specific to individual brands and product lines, not the
widely used and applied indiscriminately to a broad range company as a whole. Attempting to model at the corporate
of marketing models used to evaluate different components (or aggregate) level rarely works because what is going on
of marketing plans, such as advertising, promotion, in one part of the company is canceling out or confounding
packaging, media weight levels, sales-force numbers, what is going on elsewhere in the company. Here are
etc. These models can be of many types, but multiple some types of data to consider when developing the data
regression techniques lie at the heart of most marketing warehouse:
mix modeling. Regression is based on a number of inputs „„ Economic data. An understanding of the effects of
(or independent variables) and how these relate to an general economic variables is vital to building sound
outcome (or dependent variable), such as sales or profits, models. Some economic variables include employment
or both. Once the model is built and validated, the input and unemployment, discretionary income, inflation rates,
variables (advertising, promotion, etc.) can be manipulated gross domestic product, interest rates, and energy costs.
to determine the net effect on a company’s sales or profits.
„„ Industry data. What are the trends in the specific
If the president of a company knows that sales will go up industry? Is the market for the product or service
$10 million for every $1 million he spends on a particular growing? What is the rate of growth? Is international
advertising campaign, he can quickly determine if additional trade affecting the industry? Are important geographic
advertising investment makes economic sense. A scientific differences evident within the industry?
understanding of the variables that drive sales and profits
„„ Product category data. What are the trends in
is essential to determining an optimal strategy for the
the specific product category? For example, is the
corporation. Marketing mix modeling creates a broad
refrigerated soy milk category growing? At what rate?
platform of knowledge to guide strategic budget allocations
How does this growth rate vary by geographic region?
and decisions.
What are the trends by brand?
From a conceptual perspective, there are two main
„„ Product lines and SKUs (Stock Keeping Units). What
strategies to pursue in marketing mix modeling. One is
is the history of each major brand within the category?
longitudinal; the other is cross-sectional or side-by-side
What new products or new SKUs have been introduced,
analyses. In longitudinal analyses, the corporation looks at
and when, for each major brand? What is the history of
sales and profits over a number of time periods (months,
private-label brands and SKUs in the category?
quarters, years), compared to the marketing inputs in each
of those time periods. In the cross-sectional approach, „„ Pricing data. A history of prices for each SKU in the
the corporation’s various sales territories each receive category is essential. Pricing differences across sales
different marketing inputs at the same time, or these regions, across different time periods, and across brands
inputs are systematically varied across the sales territories in a category provide the data for developing precise
and compared to the sales and profit outcomes. Both price demand curves. Pricing is almost always an
methods are sound, and both have their place. Often, some important variable.
combination of the two methods is the most efficient. „„ Distribution levels. What is the history of distribution
levels for each product and SKU? What is the quality of
The Data Warehouse that distribution? What is the average number of shelf
The greatest barrier to successful modeling is always a facings per SKU?
lack of relevant, specific, accurate data. So, the first step in „„ Retail depletions. It’s essential to have a clean
any modeling effort is designing the data warehouse that measure of sales to end users, undistorted by
will support the modeling. The next step is collecting and fluctuations in inventories. Factory shipments are
cleaning all of the historical data and entering it into the data worthless for modeling purposes in most instances.
warehouse, and then cleaning and entering new data on Retail takeaway (or retail depletions) in dollars and in
a continuing basis. Clean, accurate, highly specific data is

2 Decision Analyst: Marketing Optimization Copyright © 2016 Decision Analyst. All rights reserved.
 units (ounces, pounds, cases, etc.) is the most common „„ Sales-force effects. Every company and industry is
measure of sales to consumers. The goal is to accurately different, but the nature and strength of a company’s
measure sales to ultimate users. sales force (and how it is organized, managed, and
„„ Advertising measures. Money spent on media compensated) have important strategic effects. Sales
advertising is seldom useful by itself. The media organizations tend to be very expensive, so it’s generally
advertising must be translated into television GRP worthwhile to include sales-force variables within the
(gross rating point) equivalents, or some other common models.
“currency.” That is, the print advertising, the radio „„ Service effects. If services are an important part of the
advertising, the online advertising, and any other customer’s experience in buying and/or using a product,
advertising must all be converted into common units then this variable must be measured and incorporated
of measure (typically, television GRP equivalents). into the models. For example, if a new product must be
The money spent by specific media type (adjusted for installed by a service technician, then the interaction
comparative effectiveness) is another way of weighting between customer and technician might be a major
media inputs as variables. All of this is apt to prove variable.
worthless, however, if copy-testing scores are not
included for each of the ads. A media plan of 100 GRPs Depending on the industry and product category, other
per week might have no effect if a weak commercial is variables might come into play as well. Every company
run, but might have great effect if a terrific commercial is and brand is unique, and identifying all of the relevant
aired. Likewise, the exact media schedule is important, variables, figuring out how to measure them, and getting
and the length of time each commercial is on the air must those variables into the data warehouse are the most
be considered because of wearout effects. difficult parts of establishing a successful modeling
program. Most importantly, the data warehouse must be
„„ Consumer promotion. Consumer (or end-user)
carefully maintained over time and be constantly updated,
promotions come in many forms, but the primary
because marketing modeling cannot be a one-time thing.
characteristic of these promotions (as compared to
The models must be calibrated and reweighted on a regular
advertising) is the immediacy of the effects. Promotions
basis, at least once a year. Many companies hire at least
are designed to have powerful, short-term effects on
one full-time employee devoted to tracking down relevant
sales. Temporary price reductions, cents-off coupons,
data, cleaning it, coding it, and getting it into the data
and “buy one/get one free” are examples of common
warehouse. Often, the analytical firm guiding the modeling
consumer promotions. These promotions must be
will place employees on site to help ensure that the data
understood, measured, and incorporated into the models.
warehouse is properly maintained.
If not fully comprehended, the promotion effects could
easily overwhelm the modeling effort.
„„ Trade promotion. These promotions usually take
Rules of Thumb
the form of discounts or allowances given to the trade Remember that the modeling must be specific to an
to stage in-store promotions of some type (temporary individual brand (or narrow line of business), because what
price reductions, end-of-aisle displays, in-store signage, works for one brand or one company might not work for
local advertising, and so on). Trade promotions must the next brand or the next company. As a company learns
be comprehended and included within the models what drives its individual brands, commonalities are often
because of the sales fluctuations they cause. When the found that make it easier, and less costly, to build marketing
manufacturer offers one dollar off the price of each case models for its other brands. Here are some rules of thumb
for 30 days (a typical trade promotion), the retailer is very to guide the modeling work:
likely to take actions to increase sales of that brand, and
to load up on inventories at the end of the promotion „„ Beware of threshold effects. Often a marketing
period. input (print advertising, for instance) never reaches a
measurable threshold and, therefore, does not show up

Copyright © 2016 Decision Analyst. All rights reserved. Decision Analyst: Marketing Optimization 3
 as important in the models. But if the print advertising’s Who should do the modeling work? Some large companies
budget had been slightly greater, perhaps it would have have internal modeling departments, but most companies
shown up as a meaningful variable (i.e., it would have will outsource the modeling and analytical work. Ideally,
reached the threshold of effectiveness). the modeling consultants should have an in-depth
„„ Be sensitive to “lagged” effects. Some marketing understanding of marketing and marketing research,
inputs have immediate effects, while the effects of other so that they really understand the complexities of the
inputs are “lagged” (that is, occur over time or occur at a marketing variables they are trying to simulate. True, the
later point in time). For example, media advertising tends model builders need statistical and mathematical skills, but
to have short-term effects on sales, as well as longer- without the marketing knowledge and marketing research
term effects. experience, the modeling effort is not likely to be very
successful.
„„ Keep it simple (at least in the beginning). Focus
the modeling efforts on a limited number of major Lastly, the issue of cost and ROME (return on modeling
marketing variables (the big budget items). Don’t clutter effort) must be considered. To set up and operate a
up the models with a large number of trivial variables comprehensive marketing mix modeling program can cost
that complicate and confound the modeling work. Once hundreds of thousands of dollars a year, or even millions
the major variables are truly understood, then smaller per year for a large, multibrand company. Does your
variables can be explored. organization have the stomach for that kind of ongoing
investment? Will your company really use the results? Will
„„ Be realistic. It may take several years of diligent effort
senior management heed the findings? Will the learnings
before the marketing mix modeling begins to pay off.
be disseminated throughout the organization to improve
There are no instant cures or short-term solutions. It is
strategic planning? Every company must ask itself these
hard work, trial and error, and a long-term search for
hard questions. If the answers are positive, and the
marketing truth.
company is willing to pursue the objective, scientific truth
„„ Seek top management commitment. Involve the about its marketing efforts, then marketing mix modeling
senior leadership of the corporation in the modeling can lead to sustained, long-term sales growth and improved
effort, especially at the initial planning stages. Their profitability.
understanding of the nitty-gritty details of the industry, the
company, and the brands will help ensure the success of
the modeling effort, and will encourage the acceptance,
dissemination, and use of the results.

About the Author

Jerry W. Thomas is the President/CEO of Decision Analyst. The author may be reached by email at
jthomas@decisionanalyst.com or by phone at 1-800-262-5974 or 1-817-640-6166.

Decision Analyst is a global marketing research and analytical consulting firm. The company specializes in
advertising testing, strategy research, new products research, and advanced modeling for marketing-decision
optimization.

604 Avenue H East  Arlington, TX 76011-3100, USA

1.817.640.6166 or 1.800. ANALYSIS  www.decisionanalyst.com

4 Decision Analyst: Marketing Optimization Copyright © 2016 Decision Analyst. All rights reserved.