A simple approach to analysing promotions is to assess their impact in terms of volume, value and profitability. This is achieved by means of estimating the base volume, i.e., the expected sales volume in the absence of short term causal influences such as promotions. The result is depicted by means of a baseline dashboard such as the one shown here, which utilizes a proprietary algorithm for estimating base volume. When information, such as this, on promotions and other causal influences is plotted alongside the baseline and sales data, you get to see the impact of the causal factors on sales.
Promotions can vary considerably in terms of impact. Some of them seem to work better than others; some items respond better to promotional influences; some promotions generate big gains, whereas others cannibalize; some promotions are profitable, others incur loss.
It is important, however, to appreciate that the impact of a promotion should not be gauged solely on the performance of an item, or a brand or even a category. Taking a blinkered view can be misleading both for the manufacturer as well as the retailer.
Consider for instance that an effective loss leader is unprofitable when view in isolation, and yet it may be highly profitable when assessed in terms of the retailers total business. For the retailer, promotions must be understood in the context of category roles and strategies, and in many instances you need to evaluate their overall impact through retail analytics and consumer analytics.
Baseline analysis is very useful because it quickly and inexpensively provides information on a large number of brands, in a manner that a layman is able to understand. It does not however, answer a number of critical questions, including the ones listed below:
Econometric promotions response modelling can provide answers to all of the questions posed above. 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.
Aside from promotions, market mix models can assess the impact of other elements of the mix, so that marketers may align their brand’s initiatives into a coordinated programme designed to drive revenue and profit. The models use statistical methods of analysis of historic market data, to estimate the impact of various marketing activities on sales. They reveal the effectiveness of the marketing mix elements in terms of their contribution to sales and profits, and can be benchmarked against costs to compute ROI. This knowledge empowers marketers to craft plans that optimize the use of resources, and infuse marketing decisions with the logic and discipline of analytics.
Details about key design considerations for market mix models are provided in the Marketing Analytics eGuide. The guide also offers guidance on how to apply, and how to interpret and analyse the output of these models. Snippets from the relevant chapter, Market Mix Modelling, provide illustrations of some of the techniques commonly applied by practitioners, including sales decomposition, due-to analysis and what-if analysis.
Market Intelligence and Data Visualization
Custom designed, interactive, automated, online dashboards that allow for the integration and visualization of market knowledge from diverse sources, in a manner that makes it easy to access and digest.
Consumer Analytics, Consumer Panels
Analysis of continuous individual/household level (customer level) behavioural data to address business issues.
Scan Track and Retail Analytics
Dashboards and analytic solutions for reading/analysing retail audit/scan track data; and techniques for analysis of continuous outlet level transaction and shopper data to address business issues.
Analytic techniques/dashboards for evaluating consumer promotions in terms of gains in volume value and profit; estimating discount price elasticity, price cross elasticity, decomposing sales, and applying due-to and what-if analysis.
Quantitative Research, Customer Satisfaction Research
Custom designed solutions for data processing and reporting with a working example of an interactive, automated online dashboard for Customer Satisfaction Research.
Ashok Charan, Associate Professor,
NUS Business School, National University of Singapore,