Gary L. Lilien.

How many dollars for industrial advertising? research in progress (ADVISOR Project) online

. (page 1 of 2)
Online LibraryGary L. LilienHow many dollars for industrial advertising? research in progress (ADVISOR Project) → online text (page 1 of 2)
Font size
QR-code for this ebook


D28

M414

735-14A



WORKING PAPER
ALFRED P. SLOAN SCHOOL OF MANAGEMENT



MASS. ir:ST. TECH.
OliC 1 3 T^

DEWEY LIBRARY




HOW MANY DOLLARS FOR INDUSTRIAL ADVERTISING?

Research in Progress
(ADVISOR* Project)



Gary L. Li lien



Revised



November 1974



73')-74 /A



MASSACHUSETTS

INSTITUTE OF TECHNOLOGY

50 MEMORIAL DRIVE

CAMBRIDGE, MASSACHUSETTS 02139



HOW MANY DOLLARS FOR INDUSTRIAL ADVERTISING?

Research in Progress
(ADVISOR* Project)



Gary L. Li lien



^wss. i^isr. 71



TECH.



iJ'^C



n '74



DtWEy ub:



Revised



November 1974



73-)-74 f\



ADVe rtising Industrial products: S^tudy of 0])erating Relationships



Abstract

Companies selling to industrial and business markets face the problem
of determining how much to spend for various elements in the marketing mix.
Setting budgets for advertising and other marketing communication expendi-
tures is especially difficult. This paper describes the ADVISOR project,
a cross sectional study of current practice in setting communications budgets
for industrial products. Anticipated managerial uses of the results are
described as are the methodology of the study and progress to date.



'975 23 JAN



IABLE_OFCONTENTS



1 n • Page

'. Project Overview

^' What •s'Tn'fE^"^'i'^^^ ^° Process or
What s In It for Marketing Management? ..

5

Appendices ^

A. Prototype Interactive Program

B. |J:,i;^-trial Advertising Budgeting Process: '"An'inU^

C. Proposed Project Methodology

•^ 15

D- Data Security Procedure

r T . ^ 16

^- lext of talk given at ARF Meetinq

^ 18



()723i^0(.



I. Project Overview

Every company selling industrial products faces the marketing-mix problem:
How should funds be allocated to such activities as direct sales, customer service,
and marketing communications (advertising and other customer-directed promotions)?
Should a given product be advertised at all? What types of communications will
best support current selling objectives? Are there special requirements, in this
market, at this time - competitive or otherwise - which indicate a need or
opportunity for changes in marketing expenditures?

Each company brings much experience and thought to setting budgets and
making plans for marketing communications. However, to a large degree, these
decisions are based on judgement and historical experience, rather than facts.
Very little quantified intelligence exists on the relation of product and market
characteristics to communications expenditures for industrial products. One
reason for this is that to conduct such studies individually would, for most
products, cost more than is economically justifiable.

We can differentiate between at least three classes of methods for allocating
communications expenditures:

a. Guideline Methods (Rule of Thumb) - These include such suggestions as
"budget a constant percent of sales," "match competition," etc.

b. Objectives (Task) Method - This uses intermediate measures of effective-
ness for evaluating communications programs and establishes cost con-
straints for various portions of total expense. It calls for explicit
thought about various issues such as position in product life-cycle,
state of the marketing environment and corporate objectives.



c. Explicit modeling and/or experimentation method - This approach
attempts to relate marketing actions to profit or other objectives
via theory and direct measurement.

None of these methods have been found to be cure-alls. Present guideline
methods fail to answer the hard questions like "What percent of sales?" or
"Why match competition — what makes us think they're right?" Task methods
introduce intermediate variables but have difficulty relating them clearly
to final measures of effectiveness. Explicit modeling and experimentation
are generally expensive. Basically, not enough is known about the sales
response to industrial communications.

Yet, on the positive side, it is obvious that a large number of marketers
have been making decisions for a long time and that in a pragmatic "survival of
the fittest" sense they have, by and large, been successful. This means they
have learned enough about their jobs to make good decisions, "on the average."
Therefore, a careful study of current practice, the goal of the ADVISOR project,
offers the possibility of developing a new fonii of guideline method. The method
focuses on systematic cross-sectional analysis of current practice and infers
decision making behavior from actual data. Although it would be presumptuous
to expect a complete theory to emerge from this study, it is hoped it will be
an important step toward answering the question posed in the title of this paper.

The goal of this study can be understood if one were asked what, given
existing knowledge, is a pragmatic way to approach the problem. A good solu-
tion is to hire an experienced man who had worked on setting communications
budgets for several hundred products with widely varying characteristics and
cormiuni cations needs. His accumulated experience would make him a valuable
resource for aid in decision making; a systematic analysis of his experience



could lead to a new understanding of the objectives and results of the com-
munications budget setting procedure. The goal of this study can be viewed
as a duplication and analysis of the accumulated knowledge of this "experienced
marketing man. "

As of this date, twelve companies are participating in the study, providing
data on 10-25 of their products, as well as project support. Table 1 lists
current participants; more companies are expected to commit to participation
before data collection is complete near the end of 1974.

Continental Can Company I.T.T.

E. I. Du Pont de Nemours & Company Monsanto Company

Emery Industries 01 in Corporation

General Electric Company Owens Corning

International Harvester Company U.S. Steel

Chase Manhattan Bank Union Carbide



TABLE 1 : Participating Companies
[As of August 15, 1974)



Data are being submitted on a range of marketing variables for each pro-
duct included. Marketing expenditures for calendar 1973 are the basis for
the study — both as a total dollar amount and broken into sub-categories in-
cluding, space advertising, direct mail, trade shows, etc. Partial data are
also being collected for the previous year to establish "changes in 1973." The
variables related to expenditures are a series of product and market charac-
teristics, including:

1. Direct sales expense for the product.

2. Industry sales of product and growth rate.

3. Company sales of product and growth rate.

4. Approx. number of 1973 company customers.

5. Approx. number of potential customers (non-buyers).

6. Geographic concentration of customers.



-4-



7. Product Return on Investment.

8. Product Return on Sales.

9. Product production capacity utilization - industry and company

10. Distribution — geographic and number of levels.

11. Position in product life cycle.

12. Product uniqueness (rating).

13. Product - company association (rating).

14. Cost significance of product to user (rating).

15. Technical service expenses.

16. Market purchasing practices (rating).

17. Opportunity for marketing communications spill -over to related pro-
ducts (rating).

These data will then be analyzed in order to ( 1 ) determine and quantify
the market and product factors which influence the decision of whether or
not to use marketing communications in promoting a given product or product
category; (2) determine and quantify the 'iiarket and product factors which
influence how much and what types of marketing commuriicatiotis arc used; (3)
establish industry norms for use by individual companies to help in determin-
ing how to allocate the communications budget among the various forms of
marketing communications.

An additional study objective is to develop sufficient understanding of
this process to be able to do initial research on the development of decision
making aids in the industrial communications area.



-5-



2. How the Output Relates to Process or What's in it for Marketing Management?

This study will identify those market and product factors which most
importantly affect marketing communications expenditures levels. An analysis
will be generated for each product included in the study, comparing its per-
formance with an "industry average" as determined by the model. This could
be used as input for individual product audits by participating companies.

An interactive computer program will be developed which will allow par-
ticipants at remote terminals to "play" with the model. A user could enter
marketing data in a conversational mode and get the results one would expect
from the "experienced marketing man" mentioned earlier, given those inputs.
This program could be of aid to participants to:

check their existing practices and possible future strategies against
the norms determined by the analysis;

develop possible spending levels for new products or those not currently
using marketing communications;

provide new input and rationale for recommending budget levels to pro-
duct marketing people.

Appendix A includes a very rough example of such a system.

A report will also be made assessing the feasibility of developing a
model which aids in deciding how much should be spent for communications.

To summarize, the research described here should be of direct use for
the product or advertising manager as well as for the general manager as follows,

For the product or advertising manager:

As changes take place in marketing goals or in the marketing environment,
guidelines are available to adjust budgets.

The list of variables is identified and available which includes the
most important factors affecting budgets.

Magnitude as well as the direction of change are provided.



-6-



■ A norm can be produced for a product or, alternatively, a set of adjust-
ments to an existing budget necessitated by market or product changes,
can be produced,

' The effects of several, simultaneous changes can be assembled separately
or jointly.

For the general manager:

' He will receive quantitative support for the budget allocation decisions
and reviews he is responsible for making.



-7-



APPENDIX A: PROTOTYPE INTERACTIVE PROGRAM

Page 8 lists the conversational input for an example of the type of
Interactive program envisioned as a result of this project. The hypothetical
product is called ENGINE EASE (user inputs are underlined); it is a chemical
product, with a 20% profit margin, is relatively new, etc.

Page 9 gives the model analysis of the communications budget for the
product. The input is first reprinted for review. Then a statement is made
telling that the characteristics of this product lead to an estimate of $25,000
as an advertising budget reference value as opposed to $13,500 for an average
product with this sales rate (one with all factors {0.} = 1).

The individual factor effects are then listed along with the product of
those factors, 1.87 in this case.

In an operational version of this model the user could then modify any
one or set of inputs and re-run the analysis.



-8-



This is AbVIoOR [-oiio (■\'!c; l'J7U). . llic r'ro-rai •ov('l(.|.s
'■'■''■' ^'"f' :(int rr.Liiatcr. U,r cor t i.n i ca t i cui expenditures
for inf'Lr_.tria] proc'ucts
inpLi t I .rf.f'iic t. lull ,e
? EliClflC L/JC



ni,i-L iiii'ui^try a ('v i.c. (i.e.=i-L



in|.Lit 7 : (.,1 cc ('e ] i ;. t. )



1 1 1 1 >■ L MP L.'^ L r y (:r indn-l
y Cl:L; tei ,e r r:



y.in.uL si.al]ei.t . ec v, rail, i c tracti(n of tir- ce



Li'i (_ I':



entry cc ve r i n; i [, '^ (.>



a' e ' c li



^.iiiluL .roctien (d' |.•'-'"•:■'• j^^rei/tl, I'Cal. 1^ eldest

V 1^

-i-". in, .Lit cei,pany posture; tC'V/arc' prci'uct as or, i ndt.;;

I— uiiii,p(>rtant prod : =civr. trcatient xf:-tu| la-ita-if.

li. inpct pree'urt uniquriess ratinp as an inc'e;;

^ C-indistlnruisI al le prer'pct i^sevcral siiilai lC-iiniML!f

• -li-

1 :: . input avr ik, cf iCH^-le nc.rrallv, invflv((' in leyin/- .iccislen •

V .^

''■'■ '■'^'■■- dnrchase dad if. ler tl is prcru.rt frei., =coi,,r 1 e te I -. stn,rtur(>d
(•T- i" M'Vt di('s) t l!.'-ir pnl^,e decisive,

J '■• • ' ''|:e t umeal cci/pany sales in nillicms ■'

i- . input anniial i r( '■pct salt-; i r. i.illiciis t

7 ' ~ ^



■9-



ADVI 30R DliMO RE3IJI.T3



liiIuL


1

Online LibraryGary L. LilienHow many dollars for industrial advertising? research in progress (ADVISOR Project) → online text (page 1 of 2)