No file available
To read the file of this research,
you can request a copy directly from the authors.
Pre-publication version to: Forecasting advertising and media effects on sales: Econometrics and alternatives
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
The contribution of regression analysis (econometrics) to advertising and media decision-making is questioned and found wanting. Econometrics cannot be expected to estimate valid and reliable forecasting models unless it is based on extensive experimental data on important variables, across varied conditions. This article canvasses alternative, evidence-based methods that have been shown to be useful for forecasting problems. These methods are described with the hope that they are more widely used for marketing forecasting. The approaches include media and copy experiments, analyses of individual level single source data, and structured expert judgment.
... ($2.8 bn.) Compared to the same period in 2018, sales increased by 15.9 percent in hryvnia and 18.3 percent in dollar terms. In physical terms, sales decreased by 3.1 percent and amounted to 1.25 billion packages. ...
... The contribution of regression analysis to media decision-making is quite significant, but there are alternative methods. Dawes et al. (2018) [15] describe evidencebased methods that have been shown to be useful for forecasting. Jin et al. (2017) [16], Zhang and Vaver (2017) [17] suggest using Bayesian hierarchical modeling. ...
... The contribution of regression analysis to media decision-making is quite significant, but there are alternative methods. Dawes et al. (2018) [15] describe evidencebased methods that have been shown to be useful for forecasting. Jin et al. (2017) [16], Zhang and Vaver (2017) [17] suggest using Bayesian hierarchical modeling. ...
Social capital has become an important aspect of most rural communities in developing nations. But, the dimensions of social capital vary across rural regions while little is known about the factors influencing it in rural areas. This study aimed to identify the prevalent social capital dimensions in rural areas and examine the factors determining rural people involved in those dimensions. A field survey which consists of structured and self-administered questionnaire was carried out with rural households. The information of the survey was obtained from 220 rural households in the study area between August and October, 2019. The descriptive analysis identified social networks (3.875), norms (societal values) (3.390), trust and solidarity (4.115), and cooperation and group action (4.139) as the prevailing social capital dimensions in the rural communities. The results further suggest that cooperation, trust and solidarity, and networks are respectively the dominating social capital dimensions in the rural areas. The results from probit model estimates show that the factors that are more likely to be associated with social capital in rural areas include education, access to credit and ownership of farm (cash crop). Since social capital is becoming a prerequisite for rural development, our findings lead to the suggestion that cooperation, build-up of networks should be facilitated for people in the rural areas. Furthermore, policy direction towards access to education, credit provision and development of primary occupation in the rural areas should also be enhanced. Economic policy makers and rural development agencies are invited to continuously work on the identified factors to promote the individual, community and national development on equitable basis.
... The contribution of regression analysis to media decision-making is quite significant, but there are alternative methods. Dawes et al. (2018) [5] describe evidence-based methods that have been shown to be useful for forecasting problems. Jin et al. (2017) [7], Zhang and Vaver (2017) [17] suggest using Bayesian hierarchical modelling. ...
... The contribution of regression analysis to media decision-making is quite significant, but there are alternative methods. Dawes et al. (2018) [5] describe evidence-based methods that have been shown to be useful for forecasting problems. Jin et al. (2017) [7], Zhang and Vaver (2017) [17] suggest using Bayesian hierarchical modelling. ...
The objective of this paper is to research, modeling and forecast the call-center workload that depends from all media and marketing activities. Data mining approach and machine learning technologies help to clearly identify and distinguish the impact factors on the feedback of potential customers (both positive and negative), determine which communication channels to use to increase inflow of queries. The model for forecasting of effectiveness of media investments and as a result managing of Return of Marketing Investments (ROMI) based on hourly data for all calls to Call Center, media and marketing indicators and macroeconomic factors for banking sector in Ukraine for the period 2013-2018 years was built. Authors used such machine learning technology as econo-metric modeling (regression analysis) for key metric "Incoming Calls to the Call Center". Data Science technologies help to forecast and manage calls flow with average error that is less than 10%. Article describes how to increase the effectiveness of advertising campaign by 8% in the first 2 months and achieve potential growth of conversion rate by 58%, compared to the standard market level. This article contains the key stages of implementing data mining approach, directly in the process of machine learning and dwell on the important technical aspects of the implementation of forecasting models.
... The contribution of classic machine learning methods like regression analysis to marketing decision-making is quite important, but there are alternative methods. J. Dawes et al. [18] research evidencebased methods that have been shown to be useful for forecasting. Y. Jin et al. [19], S. Zhang and J. Vaver [20] recommend using Bayesian hierarchical modelling. ...
... Estimation the impact of media factors by each brand, comparison of such influence by brands in terms of profitability help to develop recommendations for media strategy of these drugs, prioritize brands by media support and marketing budget allocation and as a result increase ROI of media/marketing investments for each brand separately and for portfolio in general [18]. ...
The article contains the results of Data Science realization for pharmaceutical market. The main goals of research are modelling the sales of three key brands of one of the Ukrainian pharmaceutical companies based on regression analysis as a main method and to make conclusions for portfolio marketing strategy optimization and effective brand management. Estimation the influence of key elements of the marketing mix on brand's share of market make basis for ROI calculations and optimization of media budget allocation between brands in portfolio. The article contains recommendations for analytical system construction.
... However, these methods are either proprietary, rely on expensive data, or require analytical abilities beyond the skillset of practitioners. Furthermore, while it is true that additional variables often lead to more accurate results, the validity of these results are limited to a single set of data (Ehrenberg, Barnard & Sharp, 2000), are regularly skewed from the cherry-picking of input variables, and are prone to discovering spurious relationships (Dawes, Kennedy, Green & Sharp, 2018). Therefore the models themselves provide few contributions to real-life decision making and are often ignored in favour of less accurate methods, such as the Sainsbury Normal Method (SNM) (Caffyn & Sagovsky, 1963). ...
... That is, not only where the SNM estimates are weighed by age, or by gender, or by location, but weighted by the various combinations of age, gender, and location. Although this may provide more accurate net-reach estimations, would the results, therefore, be restricted to that single situation (Ehrenberg et al., 2000), or would the requirement for marketers to select how many, and which segmentation variables limit their broader applicability (Dawes et al., 2018). ...
For almost nine decades, advertisers have relied on the Sainsbury Normal Method (SNM) to estimate net-reach where single-source data are either too expensive or unavailable. To the best of the authors' knowledge, no SNM validation studies have included catalogues, smartphone applications, websites, social media, or cinema. While few studies have applied the SNM across media, no study has addressed the limitation of the SNM, that is, the implied assumption of audience homogeneity. Given that audiences do differ by age within any medium, and across media, there is a need to incorporate audience heterogeneity into the method. The authors introduce the Sainsbury Weighted Method (SWM) which provides more accurate within medium net-reach estimations in 82% of the 9,680 cases analysed, with an average accuracy improvement for within medium net-reach estimations of 0.6 percentage points (or 13%). For across media net-reach, the SWM estimations are more accurate in 77% of the 968 cases analysed, improving the average accuracy by 0.5 percentage points (or 49%). Reach Word count 6864 Summary Statement of Contribution If within medium or across media audience profiles are dissimilar; then in the majority of cases, our new method improves net-reach estimations for catalogue, smartphone application, newspaper, television, radio, magazine, cinema, outdoor, website, and social media. The new formula (the Sainsbury Weighted Method) does this by overcoming the shortcomings of the previous formula (the Sainsbury Normal Method). The easy to calculate estimations are especially useful for brand managers and media schedulers who do not have access to single-source data which provides observed net-reach. 2
... The greatest research of the theoretical and practical issues of marketing pricing policy was found in the works of Balabanova L., Osypenko S., Romanchyk T. (Osypenko et al., 2020) The contribution of classic machine learning methods like regression analysis to marketing decisionmaking is quite important, but there are alternative methods. Dawes et al. (2018) research evidence-based methods that have been shown to be useful for forecasting. Jin et al. (2017), Zhang and Vaver (2017) recommend using Bayesian hierarchical modelling. ...
The article contains the results of Data Science technologies application (including machine learning and regression analysis) to modelling the results of marketing activities of key brand of one of the Ukrainian pharmaceutical companies on the basis of historical data for the period from 2015 to 2019 in weekly detail. The main goal of research is to estimate the influence of key elements of the marketing mix (penetration of pharmacy chains, price policy vs main competitors, advertising activity of the brand and its competitors in all communication channels (television, Digital, radio, outdoor advertising, press)) on company’s sales, volume market share and value market share in relevant segment of drugs. Based on the results obtained, the article explains in detail the impact of penetration, price policy and media activity on the competitiveness of the enterprise and its position in the market. The influence of the price policy and penetration directly on sales (market share), as well as on other factors (including the effectiveness of the brand's advertising activity on television) is estimated and taken into account for development the effective marketing strategy. Based on the research, the article contains main recommendations for optimizing the marketing strategy to maximize the company's sales and increasing market share in monetary or physical terms. Data Science technologies become a tool for sales management, because it creates the ability to quantify the impact of each factor on sales, determine their optimal combination for achievement of business goals and strengthening the company's position in the market, effective marketing budgets distribution and scenario forecasting. Continuous model support allows to increase the return on each factor, improve return on investment and ensure the achievement of business goals in the most efficient way. Data Science forms the basis for finding effective marketing solutions and forming an effective business development strategy.
Purpose
Managers engage in marketing efforts to boost sales and in setting marketing budgets based on current or historical sales. Past studies have overlooked the reciprocal relationship between marketing spending and sales. This study aims to examine the nature of the relationship between sales and marketing expenses in the B2B market.
Design/methodology/approach
Five hypotheses on the relationship between sales and marketing expenditures were framed. A total of 30 of India’s dyeing firms provided data on revenues, sales (in units) and marketing expenditures over time. The structural vector auto-regressive model and the vector error correction model were fitted to the data.
Findings
The results show that marketing expenses and sales are related bidirectionally in a sequential way. Furthermore, sales drive the long-term equilibrium relationship to a greater extent than marketing expenditures.
Practical implications
The findings of this study should assist managers in predicting sales and marketing budgets simultaneously and devising precise marketing strategies and tactics.
Originality/value
Using econometric models in data-driven research is not a frequent practice in marketing. This study adds value to the body of marketing literature by advancing the theory of the relationship between sales and marketing spending using real-world data and econometric models in the B2B sector.
The article contains the results of applying marketing mix modeling based on Data Science technologies for FMCG companies. The market share in packages (sales level) was modeled using regression analysis depending on the key elements of the marketing complex (price, place, promotion), seasonality and media activity of the competitors in all communication channels. Econometric modeling helps to assess the return of media investment by calculating the level of sales generated by media activity in each communication channel and comparing it with the level of media investment, respectively. The influence of distribution on the company’s position in the market and media efficiency has been studied in detail. There is a connection between distribution and media response: less distribution affects the decline in media performance, and vice versa. In conditions of low distribution, it is important to increase the presence in regional communication channels through media pressure in critical sales regions for FMCG brands and try to increase distribution levels nationally. The article contains an assessment of price sensitivity (elasticity) and recommendations for optimizing pricing policy to increase market share by volume or by value depending on the company’s goals. The price elasticity curve was determined by estimating the impact of the price index on the level of sales in packages and deals in money using econometric modeling and simulations of sales levels depending on different options of the price index vs competitors. Based on the research, recommendations for optimization of the marketing and media strategies to maximize sales of FMCG companies are formed. Marketing mix modeling and Data Science provide the most efficient ways to achieve business KPIs.
Agent-based models establish a suitable simulation technique to recreate real complex systems, such as those approached in marketing. Reinforcement learning is about learning a behavior policy in order to maximize a long-term reward signal. In this work, we develop a deep reinforcement learning agent that represents a brand in an agent-based model of a market. The goal of the learning agent is to obtain a marketing investment strategy that improves the awareness of its corresponding brand in the marketing scenario. In opposition to conventional marketing investment strategies, the learned strategy is dynamic, so the agent makes its investment decision on-line based on the current state of the market. We choose the Double Deep Q-Network algorithm to train this agent on diverse instances of the model, each of them with different knowledge levels of the brand. First we adjust a subset of the hyperparameters of Double Deep Q-Network on two of the model instances, and then we use the best configuration found to train the agent on all the available instances. The brand agent learns a dynamic policy that optimizes brand’s awareness levels. We perform an expert analysis of the policy obtained, where we observe that the learning brand agent tends to increase investment in media channels with greater awareness impact, but it also invests in other channels according to the situation and the characteristics of the model instance. These results show the benefits of having an on-line dynamic learning environment in a decision support system for media planning in marketing.
This article describes the history of the journal published by the UK Market research Society (MRS). The reasons for introducing a journal are discussed, and the competitive context, at that time and onwards. Changes in the format, title and content of the journal over time are described, and its evolution from an in-house production to its current status within the journal portfolio of a leading international academic publishing house. The changing nature of the two main communities with the most interest in the journal, academics and practitioners, are described and the impact on the journal are discussed. Finally, the article summarises some of the many key contributions that the journal has made to the body of knowledge and evidence in the fields of market and social research.
In recent years, machine learning models based on big data have been introduced into marketing in order to transform customer data into meaningful insights and to make strategic decisions by making more accurate predictions. Although there is a large amount of literature on demand forecasting, there is a lack of research about how marketing strategies such as advertising and other promotional activities affect demand. Therefore, an accurate demand-forecasting model can make significant academic and practical contributions for business sustainability. The purpose of this article is to evaluate machine learning methods to provide accuracy in forecasting demand based on advertising expenses. The study focuses on a prediction mechanism based on several Machine Learning techniques—Support Vector Regression (SVR), Random Forest Regression (RFR) and Decision Tree Regressor (DTR) and deep learning techniques—Artificial Neural Network (ANN), Long Short Term Memory (LSTM),—to deal with demand forecasting based on advertising expenses. Deep learning is a powerful technique that can solve marketing problems based on both classification and regression algorithms. Accordingly, a television manufacturer’s real market dataset consisting of advertising expenditures, sales and demand forecasting via chosen machine learning methods was analyzed and compared in terms of the accuracy of demand forecasting. As a result, Long Short Term Memory has been found to be superior to other models in providing highly accurate prediction results for demand forecasting based on advertising expenses.
There have been frequent calls in the literature for a more comprehensive understanding of marketing impact on long-term firm performance. Retail scanner data has been the principal source of empirical evidence in this strategic domain, but it cannot explain the behavioural shifts that underpin the sales dynamics it reports. With the availability of far larger and extended household panels, it is now possible to observe the effects of accumulating penetration on brand and category buying over many years. This type of data nevertheless presents theoretical and methodological challenges to researchers. In this article, we discuss an approach to extending established marketing theory to long-run repeat buying and then outline the inherent constraints of long-term panels. We illustrate these challenges using one-, five- and 10-year panel datasets and present a research agenda to progress explanatory theories of long-run brand building and category growth in this new but so far mostly untapped resource.
Finding the main influencers in brand indicators is a challenge for every marketing manager and researchers working in the branding investments area. How much to invest, which is the proper media channel mix or what is the influence of brand heritage are the questions of interest and which the paper responds to. Therefore, this paper is aiming to analyze the brand performance indicators in 2018 (awareness, trial and usage) for over 700 brands in Romania based on their investment on each media channel for 2014-2018 period and 2014 brand indicators. As for characterization of media investment 47 variables were retained, principal component analysis was used for reducing factors of influence. Thus, four main components were retrieved: media investment in absolute measures, main and second proportion in terms of media channel mix, and qualitative aspects of the brand. In conclusion, some multivariate regressions were built for identifying impact on each 2018 brand indicator using the four principal components and 2014 levels of the brand indicators.
Marketers' intuitions about the sales effectiveness of advertisements Advertisements vary enormously in their sales effectiveness, so choosing the more effective creative executions to air is an important marketing task. Such decisions are often made intuitively. This study assesses the intuitive predictions of a global sample of marketers regarding which television ads are more or less sales effective. The findings show that marketers' predictions were correct no more often than random chance. Multivariate analysis suggests that those with category experience and those in marketing or consumer insights roles make slightly better predictions. Aside from who makes better predictions, further research is needed on how to improve advertising decisions, including use of evidence-based decision support systems and team decision-making.
This research examines long-term loyalty change in a wide variety of FMCG categories in the UK and USA, over time periods ranging from six to thirteen years. The study uses three loyalty measures: polarization index (φ), average brand share of requirements (SCR), and average repertoire size. Analysis over 26 categories shows mixed results for the proposition that loyalty is declining. Overall, there is a very small decline in average SCR of 0.77 percent per year (0.77 of 1 percent); but no statistically significant change in polarization and repertoire size over time. Indeed while some specific categories exhibit slight loyalty declines others show small increases. Furthermore, several of the loyalty measures are negatively correlated with category purchase frequency and the number of SKUs in the category – that is, if these category factors increase in a year, loyalty declines somewhat in the year.
Cross-media campaigns are becoming a norm, yet there is a lack of knowledge on how they impact sales. This paper documents the sales response to cross-media campaigns and finds that, when online advertising is added to a television campaign, the extra reach achieved is primarily duplicated. Regularly a single television exposure stimulates sales among those exposed, with online advertising demonstrating a similar yet less consistent response. We do not find evidence of a synergy in sales impact, where the sum effect of exposure to both television and online is greater than the parts. We highlight challenges with such single-source research.
It has been claimed that the user profiles of directly competing brands seldom differ. This surprises many in marketing, leading to some doubts about the validity of the claim. In the empirical generalization tradition, the authors:
re-examine the previous claim using newer data;
consider the scope of the claim in terms of brands in emerging markets, private labels, variants, and composite segments; and
discuss potential boundary conditions.
Despite attempts by marketers to differentiate brands and provide customized features for distinct target audiences, the evidence of the current study confirms that user profiles of directly competing brands seldom differ.
This article proposes a unifying theory, or the Golden Rule, of forecasting. The Golden Rule of Forecasting is to be conservative. A conservative forecast is consistent with cumulative knowledge about the present and the past. To be conservative, forecasters must seek out and use all knowledge relevant to the problem, including knowledge of methods validated for the situation. Twenty-eight guidelines are logically deduced from the Golden Rule. A review of evidence identified 105 papers with experimental comparisons; 102 support the guidelines. Ignoring a single guideline increased forecast error by more than two-fifths on average. Ignoring the Golden Rule is likely to harm accuracy most when the situation is uncertain and complex, and when bias is likely. Non-experts who use the Golden Rule can identify dubious forecasts quickly and inexpensively. To date, ignorance of research findings, bias, sophisticated statistical procedures, and the proliferation of big data, have led forecasters to violate the Golden Rule. As a result, despite major advances in evidence-based forecasting methods, forecasting practice in many fields has failed to improve over the past half-century.
"More than anything else, however, I'm just plain envious. It's a book I wish I had the intelligence to write... Reading Sharp's critique of the cult of differentiation made me smile. And I laughed out loud at his characterisation of supposedly committed consumers as "uncaring cognitive misers"."--Marketing Week
"...marketers need to move beyond the psycho-babble and read this book... or be left hopelessly behind."--Joseph Tripodi, The Coca-Cola Company
"Until every marketer applies these learnings, there will be a competitive advantage for those who do."--Mitch Barnes,The Nielsen Company
"A scientific journey that reveals and explains with great rigour the Laws of Growth."--Bruce McColl, Mars Incorporated
"This book puts marketing's myth-makers, of which there are many, in their proper place."--Thomas Bayne, MountainView Learning
"A truly thought-provoking book."--Timothy Keiningham, IPSOS Loyalty
"The evidence in this book should make any marketer think hard about how they manage their brands."--Kevin Brennan, General Manager, Snacks and Marketing Director, Kellogg UK
"This book should be required reading on any marketing course."--Colin McDonald, the 'father' of Single-Source analysis and author of Tracking Advertising & Monitoring Brands
"There is competitive advantage here for those who understand and follow this book's lessons."--Jack Wakshlag, Chief Research Officer, Turner Broadcasting Systems, Inc.
Ideally, forecasting methods should be evaluated in the situations for which they will be used. Underlying the evaluation procedure is the need to test methods against reasonable alternatives. Evaluation consists of four steps: testing assumptions, testing data and methods, replicating outputs, and assessing outputs. Most principles for testing forecasting methods are based on commonly accepted methodological procedures, such as to prespecify criteria or to obtain a large sample of forecast errors. However, forecasters often violate such principles, even in academic studies. Some principles might be surprising, such as do not use R-square, do not use Mean Square Error, and do not use the within-sample fit of the model to select the most accurate time-series model. A checklist of 32 principles is provided to help in systematically evaluating forecasting methods.
This paper examines the relationship between price changes and customer defection levels in a “subscription”-type market, namely car insurance. Two regression models are constructed to estimate this relationship, one model for younger customers and another for older customers. The regression models closely estimate the defection rates associated with different levels of price changes. The analysis also shows that the impact of price decreases on defection rates is less than the impact of price increases, extending previous research. The paper notes that models of this type should offer true predictive ability and therefore tests the ability of the model to predict defection rates for new data. The models performed comparatively poorly in this regard, particularly for price increases. The paper concludes that multiple sets of data are needed to develop and validate predictive models.
Some simple, nonoptimized coefficients (e.g., correlation weights, equal weights) were pitted against regression in extensive prediction competitions. After drawing calibration samples from large supersets of real and synthetic data, the researchers observed which set of sample-derived coefficients made the best predictions when applied back to the superset. When adjusted R from the calibration sample was < .6, correlation weights were typically superior to regression coefficients, even if the sample contained 100 observations per predictor; unit weights were likewise superior to all methods if adjusted R was < .4. Correlation weights were generally the best method. It was concluded that regression is rarely useful for prediction in most social science contexts.
Reports that 3 principles of human judgment apply to the decisions of a graduate admissions committee. The 1st of these principles is that a linear combination of the variables considered by the committee does a better job of predicting graduate success than does the committee; the 2nd principle is that the committee's judgment may itself be represented "paramorphically" by a linear combination of these variables, and the 3rd that this paramorphic representation is superior to the committee in predicting graduate success. (42 ref.) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Judgmental bootstrapping is a type of expert system. It translates an expert=s rules into a quantitative model by regressing the expert=s forecasts against the information that he used. Bootstrapping models apply an expert=s rules consistently, and many studies have shown that decisions and predictions from bootstrapping models are similar to those from the experts. Three studies showed that bootstrapping improved the quality of production decisions in companies. To date, research on forecasting with judgmental bootstrapping has been restricted primarily to cross-sectional data, not time-series data. Studies from psychology, education, personnel, marketing, and finance, showed that bootstrapping forecasts were more accurate than forecasts made by experts using unaided judgment. They were more accurate for eight of eleven comparisons, less accurate in one, and there were two ties. The gains in accuracy were generally substantial. Bootstrapping can be useful when historical data on the variable to be forecast are lacking or of poor quality; otherwise, econometric models should be used. Bootstrapping is most appropriate for complex situations, where judgments are unreliable, and where experts= judgments have some validity. When many forecasts are needed, bootstrapping is cost-effective. If experts differ greatly in expertise, bootstrapping can allow one to draw upon the forecasts made by the best experts. Bootstrapping aids learning; it can help to identify biases in the way experts make predictions, and it can reveal how the best experts make predictions. Finally, judgmental bootstrapping offers the possibility of conducting ?experiments@ when the historical data for causal variables have not varied over time. Thus, it can serve as a supplement for econometric models.
Studies in marketing research often start with data rather than with a theory. This exploratory or inductive approach is at odds with the more preferred scientific method where the theory precedes the data in any single research study. (See, for example, the discussion by Francis, 1957)
Because exploratory research is common, however, one might argue that it is of some value. A number of researchers have claimed that the exploratory approach leads to new and useful theories. But there is also the danger that the research will produce false leads or useless theories.
An attempt is made in this paper to illustrate the dangers inherent in the exploratory approach. The question of whether the potential benefits are large enough to outweigh the dangers is left to the reader.
The authors analyze results of 389 BehaviorScan® matched household, consumer panel, split cable, real world T.V. advertising weight, and copy tests. Additionally, study sponsors—packaged goods advertisers, T.V. networks, and advertising agencies—filled out questionnaires on 140 of the tests, which could test common beliefs about how T.V. advertising works, to evaluate strategic, media, and copy variables unavailable from the BehaviorScan® results. Although some of the variables did indeed identify T.V. advertising that positively affected sales, many of the variables did not differentiate among the sales effects of different advertising treatments. For example, increasing advertising budgets in relation to competitors does not increase sales in general. However, changing brand, copy, and media strategy in categories with many purchase occasions in which in-store merchandising is low increases the likelihood of T.V. advertising positively affecting sales. The authors’ data do not show a strong relationship between standard recall and persuasion copy test measures and sales effectiveness. The data also suggest different variable formulations for choice and market response models that include advertising.
Principles of Forecasting: A Handbook for Researchers and Practitioners summarizes knowledge from experts and from empirical studies. It provides guidelines that can be applied in fields such as economics, sociology, and psychology. It applies to problems such as those in finance (How much is this company worth?), marketing (Will a new product be successful?), personnel (How can we identify the best job candidates?), and production (What level of inventories should be kept?).
The book is edited by Professor J. Scott Armstrong of the Wharton School, University of Pennsylvania. Contributions were written by 40 leading experts in forecasting, and the 30 chapters cover all types of forecasting methods. There are judgmental methods such as Delphi, role-playing, and intentions studies. Quantitative methods include econometric methods, expert systems, and extrapolation. Some methods, such as conjoint analysis, analogies, and rule-based forecasting, integrate quantitative and judgmental procedures. In each area, the authors identify what is known in the form of `if-then principles', and they summarize evidence on these principles.
The project, developed over a four-year period, represents the first book to summarize all that is known about forecasting and to present it so that it can be used by researchers and practitioners. To ensure that the principles are correct, the authors reviewed one another's papers. In addition, external reviews were provided by more than 120 experts, some of whom reviewed many of the papers. The book includes the first comprehensive forecasting dictionary.
A new fashion in media planning, "continuity scheduling," appears to be sweeping the board in place of "effective frequency." John Philip Jones with his STAS measure has given it a large fillip. Subsequent analyses of the Adlab data from the United Kingdom and other sources broadly support this but suggest caution: in particular, STAS does not merely reflect advertising but varies with other factors peculiar to each brand. There is no substitute for knowing your own brand: an off-the-peg media solution may not be the best fit.
Recent single-source analyses have shed new light on the media construct "effective frequency." These contingency table investigations establish a relation between recent advertising exposure and brand choice, which can be misleading. Rich single-source data demands multivariate techniques that include other key variables like weight of category purchases and viewing, category seasonality, brand loyalty, competitors' advertising, advertising decay rates, and response to brand advertising. The authors' analyses of single-source data suggest that other considerations-the decay rate of advertising effects, market and media cost seasonality-should have equal, if not greater importance, than the response function or effective frequency in scheduling television campaigns.
The authors analyze results of 389 BehaviorScan® matched household, consumer panel, split cable, real world T.V. advertising weight, and copy tests. Additionally, study sponsors-packaged goods advertisers, T.V. networks, and advertising agencies-filled out questionnaires on 140 of the tests, which could test common beliefs about how T.V. advertising works, to evaluate strategic, media, and copy variables unavailable from the BehaviorScan® results. Although some of the variables did indeed identify T.V. advertising that positively affected sales, many of the variables did not differentiate among the sales effects of different advertising treatments. For example, increasing advertising budgets in relation to competitors does not increase sales in general. However, changing brand, copy, and media strategy in categories with many purchase occasions in which in-store merchandising is low increases the likelihood of T.V. advertising positively affecting sales. The authors' data do not show a strong relationship between standard recall and persuasion copy test measures and sales effectiveness. The data also suggest different variable formulations for choice and market response models that include advertising.
A.S.C. Ehrenberg first noticed and S. Weisberg then formalized a property of pairwise regression to keep its quality almost at the same level of precision while the coefficients of the model could vary over a wide span of values. This paper generalizes the estimates of the percent change in the residual standard deviation to the case of competing multiple regressions. It shows that in contrast to the simple pairwise model, the coefficients of multiple regression can be changed over a wider range of the values including the opposite by signs coefficients. Consideration of these features facilitates better understanding the properties of regression and opens a possibility to modify the obtained regression coefficients into meaningful and interpretable values using additional criteria. Several competing modifications of the linear regression with interpretable coefficients are described and compared in the Ehrenberg-Weisberg approach.
Marketing is an important area of management activity in any organisation. It generates trade and involves analysing, planning, managing and controlling activities concerned with creating and maintaining high levels of customer service and satisfaction. The marketer's central task is to make the brand easy to buy and this requires ensuring people can find it and know about it. This book does that as it covers the main concepts and principles that underlie marketing theory and practice. Bridging academic theory and real-world marketing knowledge, the book introduces students to the core topics necessary for their undergraduate studies and is designed with the future professional in mind. It clearly illustrates how marketing problems have been solved in business - connecting theory to practice. Combined with an enriched digital ebook version of the book (registration code and website included on print book cover flap) it is very practical in orientation and provides a more realistic view of marketing issues. Written by a combination of marketing academics and marketing scientists who engage with industry it presents information that is practical and interesting in a style that is theoretical and accessible.
Extensive work with single-source data since the 1960s has consistently shown that advertising has a pronounced short-term effect on sales, that this effect decays over time, and that creative copy is the largest contributor toward effectiveness. This article shares the foundations for these generalizations as well as more current examples that use Project Apollo data.
In an increasingly competitive industry, tourism managers are faced with the necessity of estimating future values of demand in the short term despite the limitations of scarcity, volatility and uncertainty. A convenient and flexible approach such as judgemental forecasting holds promise in addressing the major issues in the field of tourism demand forecasting. This paper presents an innovative approach, which uses the opportunities offered by decision support systems to tackle the main issues associated with judgemental forecasting. A forecasting system that supports collaborative short-term forecasting tasks among tourism managers is offered as a case example.
In this age of rapid technological innovation, firms that do not stay abreast of the latest advancements in science and technology (S&T) stand a greater chance of missing opportunities than firms that maintain vigilance over the ever-changing technical environment. As a result, a resurgence of interest in technical intelligence for business is occurring in companies around the globe. Many firms now have formal technical intelligence programs to gather, analyze and use S&T information to watch their competitors, to track emerging trends in technological development and to anticipate significant technology-based changes in key markets. Careful management of technical information that affects a business can have a vital influence on corporate profitability and long-term health. This paper describes the main features of technical intelligence operations in business, drawing on the experience of several companies that develop and use intelligence information. The steps of gathering, analysing, evaluating and using information for business decisions are described and examples are given to illustrate how intelligence concepts are implemented in firms from several different industries. Practical issues such as understanding user needs, data collection, effective analysis methods and using intelligence results are covered in the paper.
Managerial decision making in marketing is the heart of the field. Strangely enough, academic work on this topic is scarce. Existing work on marketing decision making is either descriptive or takes an optimization approach, with the role of the marketing decision maker practically disappearing. There are excellent prospects for improvement, especially considering the recent work in behavioral decision making. Relevant topics are the dual-process model of decision making, learning, emotions, and expertise. We use this work to formulate interesting and relevant research questions about marketing decision making. There has also been significant progress in the methodologies for answering these questions; for example, better ways to monitor actual decision making and sophisticated behavioral laboratories and brain imaging methods.
This paper contrasts a classic example of a logit decision model with a widely used descriptive model, the Dirichlet.Decision modeling, reviewed by Leeflang and Wittink in this issue of IJRM, aims to help make marketing-mix decisions. However, we have serious doubts about this sort of modeling: its inputs, its outputs, its assumed causality, and its frequent lack of empirically grounded predictability. It also seems to seldom really take account of already well-established marketing knowledge.In contrast, descriptive modeling more simply aims to depict actual or potential marketing knowledge, and to apply it. Such modeling often deals with marketing-mix factors separately instead of attempting to do so in one overall model.
The intelligence, measurement, knowledge, models, and desktop best practice tools discussed in this article are the types of products being developed by the 21st-century business researchers who are determined to add quantifiable value to the business enterprise and the fact-based support being used by the brand and agency teams that are determined to win in the marketplace, quarter-to-quarter and year-to-year. By accounting for, improving, and achieving a return on advertising investments consistent with quarterly business objectives, what is traditionally viewed as a cost of doing business can be transformed to wise investments in the business.
Econometrics: Get the Best from Econometric Modelling
- L Cook
Cook, L. (2014) Econometrics: Get the Best from Econometric Modelling. In Admap, United
Kingdom: Warc, pp.1-6.
Always Be Testing: The Complete Guide to Google Website Optimizer
- B Eisenberg
- J Quarto-Vontivadar
- L T Davis
Eisenberg, B., Quarto-vonTivadar, J., and Davis, L.T. (2009) Always Be Testing: The
Complete Guide to Google Website Optimizer. United States of America: John Wiley & Sons.
Two Views of Tv Scheduling -How Far Apart?
- E Ephron
- S Broadbent
Ephron, E. and Broadbent, S. (1999) Two Views of Tv Scheduling -How Far Apart? Admap,
January.
Digital giants are “Weaponising” attribution and it’s driving short-termism
- A Hickman
Hickman, A. (2018) Digital Giants Are 'Weaponising' Attribution and It's Driving Short-Termism. Available at: http://www.adnews.com.au/news/digital-giants-are-weaponisingattribution-and-it-s-driving-short-termism.
P&G shifts marketing-mix biz to Nielsen, Demandtec for Faster Roi Reads
- J Neff
Neff, J. (2011) P&G Shifts Marketing-Mix Biz to Nielsen, Demandtec for Faster Roi Reads
Available.
at Last, Long Term Ad Effectiveness Measurement, the Single Source Solution
- D Poltrack
- J Doud
- L Wood
Poltrack, D., Doud, J., and Wood, L. (2014) at Last, Long Term Ad Effectiveness
Measurement, the Single Source Solution. In Audience Measurement 2014, New York: The
Advertising Research Foundation.
How Diageo is emphasising data in marketing decisions
- K Mcquater
McQuater, K. (2018) How Diageo Is Emphasising Data in Marketing Decisions. Available at:
https://www.research-live.com/article/news/how-diageo-is-emphasising-data-in-marketingdecisions/id/5035932.
Marketing Mix Modeling on Trial
- P Moriarty
- J Joseph
Moriarty, P. and Joseph, J. (2013) Marketing Mix Modeling on Trial. Chicago: IRI, pp.1-4.
Discovering the long term effects of your advertising
- J Webb
Webb, J. (2013) Discovering the Long Term Effects of Your Advertising. Available at:
https://www.huffingtonpost.co.uk/jon-webb/marketing-long-term-effects_b_4355171.html.
Two views of TV scheduling: How far apart? Admap Magazine
- E Ephron
- S Broadbent