In the age of analytics, The Marketing Analytics Practitioner’s Guide serves as a comprehensive guide to marketing management, covering the underlying concepts and their application.
As advances in technology transform the very nature of marketing, there has never been greater need for marketers to learn marketing.
Essentially a practitioner’s guide to marketing management in the 21st century, the guide blends the art and the science of marketing to reflect how the discipline has matured in the age of analytics.
Application oriented, it imparts an understanding of how to interpret market intelligence and use analytics and marketing research for taking day-to-day marketing decisions, and for developing and executing marketing strategies.
The promoted sales volume for a majority of FMCG brands accounts for a very substantial proportion of their total sales. In New Zealand, for instance, which is one of the most heavily promoted markets in the world, according to Nielsen, nearly 60% of grocery purchases in 2012 were promoted products.
But how effective are these promotions? Considering that the proportion of on-promotion sales are so high, one would imagine that companies can barely afford not to understand their impact.
This chapter dwells on the different forms of trade and consumer promotions, the need to rationalize them, and the metrics and methods to evaluate them. It explains the basic approach to assessing promotions in terms of gain in volume, value and profit. It introduces the topic of market modelling for promotions evaluation, and outlines the basic design of models. These econometric based methods, however, are covered in greater depth in the next chapter Market Mix Modelling.
Recurring promotions encourage consumers to lie in wait for attractive deals, and maintain high pantry stock. Highly attractive promotions entice consumers to stockpile for future consumption resulting in post promotion losses — i.e. dip in baseline or regular sales in the weeks immediately succeeding the promotions. Baseline sales are also affected by the intensity of promotional activities. For instance, reflected in the Exhibit shown above is a declining trend in the baseline as the intensity of promotions increases over the course of the year.
Losses on account of cannibalization compel competing brands to strike back. As competitors get drawn into the battle ground of promotions, brands retaliate in quick succession creating commotion in the market place.
The growing incidence of promotions heightens their awareness in people’s minds, and induces brand switching. The resulting erosion of brand loyalty leads to the demotion of brands. And as brands vitiate, the market heads towards commoditization.
Theoretically if no brand promotes it is a "win–win". In reality however new brands need to promote to induce trial, and established brands promote to continue to attract new or lapsed consumers as well as retain and reward existing consumers. If they do not promote, they land on the "lose" side of the fence in a "win–lose" situation. And so the vicious circle of promotion–commotion–demotion becomes the reality of a competitive marketplace.... less
The basic assessment of sales promotions shown above reveals the impact of the promotions on sales, revenue and profitability. In this Exhibit, each of the three promotions has a different impact, and plays a different role. The promotion in week 46 (Case III) was a tactical promotion run by a local retailer during the week that a major global hypermarket chain entered into this market with the opening of their flagship store. Price was slashed by 18%, volume soared 5.4 times, yet at that level of discount, profits were wiped out.
The Case I promotion on the other hand is profitable and may therefore be repeated more frequency. It yields a healthy 78% increase in volume with a 5% reduction in price.
While this has not been considered in the above analysis, do note that the loss in margin due to the discount and other related costs is borne by both the retailer and the manufacturer. Manufacturers support promotions through trade incentives such as promotional funds, and discounts on their list price, which are usually quantity based.
From the retailer’s perspective, the analysis must also account for the increase in store traffic, resulting from the promotion. This is likely to be significant for Case III, where the price discount is as high as 18%. If the item is a traffic builder, profits from the increase in sales of all the other products bought by the gained shoppers, could easily offset the loss in profits incurred by the item.
A basic assessment of sales promotions reveals the impact of the promotions on sales, revenue and profitability. It does not however answer a number of critical questions, including the ones listed below:
Econometric promotions response models yield answers to the first five questions posed above, and the remaining question can be addressed by retail analytics.
The outline of the econometric promotions models, in terms of inputs and outputs, is shown in the Exhibit below. These models analyse data to establish the impact of each individual element of a promotion on sales. The sales response functions derived from these models yield estimates of discount elasticity of demand, discount cross elasticity of demand, and sales lifts due to displays, co-op advertising and other causal factors. It is possible to decompose sales into all of the elements contributing to the volume. Promotion response models can also forecast what impact a possible combination of initiatives will have on sales. These econometric modelling methods are covered in some detail in the chapter Market Mix Modelling.
Experienced market modellers will concur that it is easy to construct market models that are visually impressive, where the predicted and actual data match closely (as seen above), the R-square value is high, and yet the model is invalid or even nonsensical. As a user of market models, you therefore need to be a reasonably good judge of the quality of a market model.
First and foremost, when it comes to developing market models, the knowledge of the market is as important as the knowledge of econometrics. The decision maker who uses the model and the econometrician, who builds it, need to work closely to create a practical solution based on market realities. It is very important that the market dynamics are clearly understood by the developer, that all of the variables that drive performance are included.
All too often in an era of commoditization of market modelling, data is shipped from the marketer to the market modeller, without the necessary information about the characteristics or nuances of the market. A modeller based overseas may have no knowledge of the Hungry Ghost festival, the exclusion of which may result in spuriously high discount elasticities for some FMCG brands in Singapore.
The exclusion from a model of any factor that significantly influences performance is likely to compromise the validity of the model. Unfortunately measures like R-square will still look good despite the omission. This is because marketing initiatives often occur concurrently, so the impact of the missing variables is attributed to other variables that comprise the model. The point to note is that all factors that significantly influence the dependent variable (sales), including external exogenous factors, should be included, irrespective of whether or not they are key to the objective of the research.
Greater levels of disaggregation provide for more robust and reliable market models. If 500 stores across 104 weeks (2 years) are modelled at store level, this yields 52,000 individual observations i.e. this gives us very many degrees of freedom. Moreover, by modelling each store individually the modeller is able to cut through noise, and isolate and measure the impact of price and promotional activities at store level.
When store level data is not available or accessible, modellers need to work with chain or channel level data, which introduces inaccuracies due to variations at the store level. Despite these imprecisions, chain/channel level data yields useful, fairly reliable models.
As regards the accuracy of the raw data from retailers, this is less of an issue now that relatively clean, weekly store level point-of-sale (POS) scan data is readily available in most markets.
Modelling works by correlating fluctuations in sales to those in the explanatory factors. If in the data there does not exist any variation in the movement of a factor, its potential effect is not calculable. (You cannot compute the discount elasticity of demand, if the product was not offered on discount).
Note also that when two or more factors always occur simultaneously and in similar proportions, it is not possible to untangle their individual influences.... less
Market Mix Models decompose sales into the baseline and each of the factors driving sales. For instance, the above Exhibit, which pertains to a promotions evaluation analysis, depicts the incremental gains arising from price discounts, banded packs, displays, cooperative advertising and competitive effects.
‘Due-to’ analysis such as the one shown above reveals the impact of each element of the marketing mix on the year-on-year sales. The chart reveals the incremental shifts in volume due to causal factors as well as the shifts in base volume due to factors such as regular price, advertising and distribution.
The output of a market model is a set of equations that spell out the relationship between sales and the variables of the market mix. With these equations one can predict what will happen if changes are made to the mix variables. The above Exhibit, for instance, provides an example of a “What-if” analysis tool to simulate the impact of price discounting. By changing the discounts in “%Price Off” column the user is able to see how those discounts will affect the sales volume and sales value, for the items in the category.... less