You are on page 1of 29
BRANDAID: A Marketing-Mix Model, Part 1: Structure John D.C. Little Operations Research, Vol. 23, No. 4 (Jul. - Aug., 1975), 628-655. Stable URL hitp:/ 2923%3A4%3C628%3ABAMMPI%3E2.0.CO%3B2-A Operations Research is currently published by INFORMS. Your use of the ISTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at hup:/www,jstororglabout/terms.hml. ISTOR’s Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at hupulwww.jstor-org/jourals/informs. html, ch copy of any part of'a JSTOR transmission must contain the same copyright notice that appears on the sereen or printed page of such transmission, ISTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact support @ hupulwww Fri Jul 29 22:32:57 2005 (OPERATIONS RESEARCH, Vl 9, No.4 Juy-Aapuet 175 BRANDAID: A Marketing-Mix Model, Part 1: Structure JOHN D. C, LITLE Masvachusette Insitute of Technology, Cambridge, Massachusetts (Received original November 23,1973; final, December 13, 1974) ‘Marketing managers make decisions about price, advertising, promo- tion, and other marketing variables on the basis of factual date, judg- ments, and ossumptions about how the market works. BRANDAID is © flexible, on-line model for assembling these elements to describe the morket and evaluate strategies. This paper motivates the model cond presents its mathematics. The structure is modular so that in- ividval decision areas can be added or deleted at will. The model is of the aggregate response type, in which decision variables relate closely to specific sales performance measures. The major submodels, core advertising, promotion, price, salesmen, and retail distribution. ‘The advertising submodel employs a long-run sales response to ad- vertising function and a: linear lag process. Promotional effects are built up from a characteristic time pattern for the type of promotion cond @ response curve. Solesman affect sales through @ response process structurally similar to that for advertising. Retail distribution Variables are intermediaries that the company offects and that in tum affect customer response. Submodel outputs combine muliplica- tively. Competition enters in a modular, symmetric way through @ matrix of competitive coefficients that determine the source of soles for each brand as it seeks to increase its market position A MARKETING manager bases his planning and operation decisions on a heterogeneous collection of data, judgments, and assumptions about how the market works. ‘Today's technology in computers and man. agement science should make him more produetive by organizing this ma- terial and delivering it to him in the right form at the right time, ‘Toward this end, we present an on-line marketing mix model called BRaxparp. ‘The model provides a structure for relating brand sales and profit to the ‘manager's potential actions so that he and his staff ean quickly and easily analyze strategies. In a specific application the model is customized and calibrated in terms of the marketing problems at hand, ‘The preparation of the annual marketing plan for an established consumer product often proceeds as follows. First, a planning group makes sales forecast, usually by extrapolating past trends and adjusting them judg- ‘mentally for marketing information obtained from the field, Then the on A Morketing-Mix Model, Port 1 629 financial staff ealeulates production, materials, and overhead costs on the basis ofthe forecasted sales. The difference between sales dollars and cost is the gross contribution of the product. After aggrogation across prod- ‘ucts, these funds are divided into marketing budgets, investment-like items, such as now products and plant improvement, and earnings. A jostling then takes place among internal advoeates of each use of the funds until an allocation is achieved that is not too uncomfortable. Historical prece- dents and rules of thumb such as ‘so many dollars/ease’or a fixed percent ‘age of sales strongly influence the final marketing budgets. From an outsider's point of view, a remarkable feature of the process is that it does not formally acknowledge that marketing affeets sales. Mar- keting budgets appear as a consequence of the sales foreeast, not a cause. Equally surprising, profit comes as @ decision, not a result. Sometimes management will aise the sales forecast to put more money into the plan without any change in marketing funds. Sometimes price will be increased while the units forecasted to be sold are held constant. We should understand why the system works the way it does. In the first place the company does not know in precise terms how marketing affects sales. Yet management must exereise control over the operation. Plans must be made and budgets set so that the various parts of the eom- pany ean operate, A reasonable assumption is that, if the company allo- cates to marketing next year about what it did last year, the results will be similar. Changes can be made incrementally. For established products this assumption can be the rationale for a reasonably satislactory opera~ tion. Clearly, an opportunity exists for doing the job botter. It should be possible to make conditional forecasts, in which projected sales levels de- pend on the marketing actions taken. ‘The best current information on hhow the market works can be marshalled in easy-to-use form through mod- cls, Then a variety of alternatives ean be explored and a more efficient. allocation of resources worked out. ‘Another point to understand is that the present system is not just pre dicting the future but, perhaps more important, is setting goals for the organization. Forecasts are meant to be self-fulfilling prophecies. Notice that the goal-setting function appears to be in eoniict with using a model, because a model seems to predict fatalistically that, if the company spends so-and-so much money on marketing, such-and-such profit will result. On closer examination» model is obviously not all that automatic. Just because a given media efficieney or promotional boost is planned does not, mean that it will occur. What a model does is relate an overall sales or profit goal back to individual assumptions about. performance in various subareas. It then offers an opportunity to pinpoint subgoals that ean be put together to achieve a desired overall goal.