MULTIPLE CRITERIA Assignment

Developing a Multiple Criteria Model for Budgeting

The Kemper Corporation is a large, diversified manufacturing concern which operates primarily in consumer goods areas. One of its divisions, Solonex, Inc., manufactures and distributes a liquid laundry detergent nationally under the brand name of Solclean.

The Solclean brand was first developed and marketed in 1979; it performed excellently in subsequent years with growth rates between three and seven percent in each of those years. Unfortunately, since 2004 the sales (units) have not grown but have declined. This decline became so precipitous by 2008 that the brand manager on Solclean determined that action of a dramatic nature would be required to save the detergent. As a result of information supplied to the brand manager by the research department of the advertising agency working on the Solclean account,KTL, Inc., the brand manager became convinced that the decline in Solclean sales could be attributed primarily to the increase in popularity of less expensive liquid laundry detergents.

After four months of research and planning on this problem, the brand manager on Solclean, as well as the top management at Solonex and Kemper, decided to scrap the Solclean brand altogether and develop and market a replacement brand to fight lower consumer price products head on. The product developed was to be named “Solo Liquid Laundry Detergent” and was to have considerably lower quality ingredients than Solclean. Testing showed the newly developed product performed satisfactorily despite the reduction in quality of ingredients in its formulation. In fact, KTL found, in blind tests on consumers from the target market, that 54.9 percent of all those tested found the actual, in-home performance of Solo to be "excellent." The brand manager determined that a price of $1.39 per 64 fluid ounces (1.89 liters) would compete satisfactorily in price range with most generic competitors; accordingly, all planning for introduction of Solo was to proceed on the assumption that $22.68/unit would be the price for Solo(A unit contains 12-64 ounce containers of Solo and is called a "case.") This is the price charged to distributors. With respect to consumer prices, analysis by the brand manager showed, on a relative basis, that Solo would have a retail price, on average, which was 81 percent that of the low-cost competitors’ prices.

A package was developed for Solo which was similar in appearance, though somewhat more complicated, to generic packaging. The average of ratings on a 1-10 subjective scale of four account managers at KTL of the Solo package was 6.8 points. The brand manager had succeeded in obtaining firm commitments for 43 percent of the possible retail outlets which could carry liquid detergent.

With respect to market size, recent studies showed the brand manager that 77.4 percent of all U.S. consumers use the product category at least one time during a given year. The MRD people at Solonex provided a sales projection for fiscal 2008 of 14.335 million units for the liquid detergent industry as a whole. Consumer purchasing frequency was 2.1 bottles/quarter.

A rough, though useable, estimate of the production (cost of goods sold) costs was made by the production people for the brand manager; they gave $13.61/unit as the cost based upon the specifications for the product provided by the brand manager. For profit projection purposes, administrative and other marketing (excluding advertising) costs were projected at $22.731 million for Solo in the first year (fiscal 2008); thereafter, it was expected these would grow at an annual rate of 3 percent.

It was the policy of Kemper, the parent company, to require both short-term and long-term projections of the profitability of any new product. Since Solo was considered to he a new product, the brand manager was required to undertake such analyses before the final "OK" would be given by Kemper for the specifics of the introductory plan. For the short-term aspects of planning (thirteen weeks or one quarter of the fiscal year), the brand manager decided to develop a new-product introduction model similar to that utilized by the N.W. Ayer agency; this would be done in conjunction with KTL since considerable information would be required which KTL could provide if it chose to do so. Under an agreement which would give KTL rights to use the resulting model for other clients, the brand manager obtained access to the data required for building the short-term model from KTL.

In the second phase of planning, long-term planning ( a horizon year of three years was required in such planning by Kemper), the brand manager decided to project profits (based upon sales in units projections) over three years and calculate the Net Present Value of the proposed introductory plan over the three-year horizon. The manager determined that the short-term model and the long-term projections and calculations could be joined in a single model to be installed online as a Java applet.

Short-Term Planning Model Phase

After some reference to the type of model N.W. Ayer had developed and published in the Journal of Marketing Research, the brand manager developed the following "theory" of how the market response might operate in the introduction of new brands:

BA = f(DAR,AF)

IP = (f(BA,DN,PK,FB,CU)

RR = f(IP,PS,PF,RP)

MS = f(RR)

In the first equation above, the brand manager believed that "brand awareness" could be explained by "day-after recall" scores for commercials utilized to introduce the new brand in the first thirteen weeks and the "average frequency" of the media schedule used to introduce the brand in the first thirteen weeks.

In equation (2) above, the manager believed that "initial purchase rate" at the end of thirteen weeks could be explained by "brand awareness," "distribution," "quality of packaging," whether or not the brand carried the "family brand" name, and the percentage of "consumers using" the product category.

Equation (3) above is the repeat purchase response function. It states that "repeat purchase" is a function of "initial purchase" rate, percent of target market consumers who have excellent "product satisfaction," the "purchase frequency" of consumers for the product, and the "relative price" of the new product compared to existing products.

Equation (4) is actually a semi-long term function which says that "market share" for the new product at the end of the first year on the market can be explained solely by the levels of repeat purchase rates at the end of the first thirteen weeks.

The brand manager realized that the above was all "theory" and would have to be tested on specific data; in addition, specific functional relations to replace the abstract statements in (1) through (4) would need to be developed. Both of these tasks would be conducted in the data analysis stage of the project.

The next step involved obtaining data on each of the variables noted in (1) through (4). Fortunately, KTL was able to provide the brand manager with complete information (either from its files on the new-product introductions in the past or from "expert" judgment derived by four account managers who had dealt with many new products at KTL) on all 13 variables in the four-equation response function for 18 separate new products which it had worked on for many different clients over the previous three years. Some of the products had become successful while others had failed miserably; this was desired in such data so that variation between "good" and "bad" results could be understood. The KTL data are shown in Table 1. It was noted that all brands involved in the analysis should be convenience goods, i.e., products which are purchased relatively frequently and carry a small price tag compared to "shopping goods."

The brand manager noted that if the above model could be satisfactorily built or parameterized, the advertising input required in equation (1) was some estimate of DAR scores for the commercials to be used in introduction as well as the average frequency of the media schedule to be used in the introduction. The manager understood that once the average frequency was fixed for the given target market of 62 million households, the total cost of a media schedule consisting of plausible vehicles providing a reach to all HH's in the target market could be determined. Looking back at some historical information on the various "cleaner" products sold by Kemper and assigned to KTL, the brand manager found ten instances where the market size was about 60 million HH's and the first-thirteen week average frequency and DAR scores were available (all media schedule costs were updated to reflect 2008 costs):

Cleaner Products’ First 13-Week DAR and Frequency

*(Costs updated and projected to one-year cost)

The brand manager decided that the budget figures above provided a plausible range within which it was possible to spend on advertising for Solo in the first year of introduction. At minimum, the above should certainly show the differences between spending a small amount on advertising versus spending a relatively large amount. The manager also believed KTL could develop commercials for Solo which could meet or exceed the DAR scores shown above; if budget #10 of $5.89 million was used for the introduction, for example, then a DAR score for commercials used in the first quarter would need to yield a score of at least 9.4 percent accurate recall if the projections resulting from the model could be expected to hold true.

The brand manager decided to used 16 of the 18 brands given in Table 1 to calculate the model parameters; the remaining two brands, Purex and Reef, would be used to perform an "acid test" or a cross-validation test. This would give some idea as to the performance accuracy which might be expected from the four-equation, short-term model's response function.

Long-Term Planning Model Phase

The brand manager understood, of course, that once market share was predicted by the short-term model, the sales in units for Solo could be projected by multiplying this share by the projected industry sales for liquid laundry detergent provided by the MRD people earlier. Price/unit times the sales in units would provide the basis, then, for the development of projected net profit statements in the following format:

Sales $
c.g.s _______________
GM
Admin/Other

Depreciation
Mkt g
Adv _______________
NP

Since the cost per unit projections were provided earlier by the production people who would work on Solo, cost of goods sold could easily be calculated for the above statement. With respect to the horizon year required to be used by Kemper divisions, three years, the brand manager decided to go ahead by assuming, given the best available information and judgment that sales in units would grow at a rate of five percent in years two and three. Also, the manager decided to assume that unit costs and unit prices would remain constant over all years of the planning period.

Depreciation would be calculated for each of the three years in the planning period using the "straight-line method." The manager followed the policy of Kemper in assuming the residual value of the Solo plant would be 50 percent of its value at the beginning of the planning period.

While the above profit projections would include only current expenses of the Solo operation over the three-year horizon, the calculation of the net present value, as required by Kemper of its divisions engaging in new ventures, of the project would include the capital outlays required for its implementation. In consultation with the production planning people, the brand manager determined that it would cost $2.50 per unit of capacity to remodel and retool the existing Solclean plant. The capacity figure upon which the plant cost at $2.50/unit would be estimated would be that calculated for the horizon year (year three) assuming the 5 percent annual growth rate in unit sales. It was the current discount rate of approximately 4 percent which Kemper was requiring its divisions to use in the calculation of NPV's.

Using the combined short-term and long-term planning model, what advertising expenditure among the possibilities indicated above (for the ten cleaner examples) should be recommended to Solonex and Kemper top managements?

Table 1

Brand	BA	DAR	AF	IP	DN	PK	FB	CU	RP	PS	PF	RR	MS
Spaghetti-O's	42.7	25.6	8.6	25.4	92	6.4	0	31.2	66.4	43.1	8.6	15.1	16.8
Knox Flavored	5.1	10	2	3.1	12	8.1	1	18.2	14.1	54	4.3	8.1	2.1
Bounty Chili	5.2	11	2.1	4.2	20	6	0	21.2	3.3	16.8	2.1	8.1	3.4
Cool Whip new	51.1	43.4	9.8	49.4	80	9.5	0	85.6	29.3	42.6	2.8	15.2	19.3
Gaines new	49.9	32.1	11.8	40.1	87	7.8	1	32.3	34.8	61.4	6.4	16	16.9
Del Monte New Fruit	6.1	12.2	2.3	6.7	3	3.7	1	38.4	16	42.8	2.2	8.4	6.8
Hunt's Flavored	7.6	13.9	2.7	8.3	56	2.8	1	49.1	14.4	54.9	3.6	9.7	5
Phase III	48.8	34.3	9.9	21.2	74	3.3	0	46.1	33.6	54.6	2.1	14.3	16.8
Petal Soft Soap	9.7	12.8	4	16	43	9.6	0	50	30	64.2	4.3	11.1	10.5
Nestle's Q	6.7	11.8	3.8	10.9	73	3.1	1	50.1	29.3	62.1	2.4	10.1	8
Kraft Noodle	8.1	25.4	3.9	22.1	81	4.8	1	50.1	34.7	80.1	4.3	15.2	16.4
Viva	40.7	24.2	11.6	42.6	96	6.8	0	95.4	29.9	76.8	12	15.9	16.6
Start	9.3	11.1	9.9	33.4	67	5.3	0	77.2	22.1	84.1	9.4	14	16
Palmolive new	35.3	20	9	35.3	94	7.4	1	77.4	33.3	64.2	8.6	15.2	19.4
New Handi-wipes	11.1	11.8	9.9	12.4	89	5	0	32.4	14.5	17	2.8	10.2	8.1
Kraft Sandwich	30.4	24.6	6.9	20.1	89	6.7	1	54	29.1	76	3	11.2	10.3
Purex	19.8	26	5.8	18.9	66	7.5	1	51.2	50.3	80.1	3.1	12	12.7
Reef	20.8	27.1	5.9	15.2	62	9.4	0	21.4	18.6	46.4	2.4	15.1	17

Find above data set here