MARKETING ANALYTICS

Marketing Analytics

The marketing analytics project is designed to develop analytic skills for marketing professionals who want to understand the numbers driving their business or organization. This project developed from my interests in marketing and performance measurement. Many of the skills necessary can be achieved with some basic knowledge about numbers and arithmetic applied to specific marketing activities. Depending on budgets organizations may decided to develop and build bespoke dashboards while others may decide to use already available software such as Microsoft Excel. Not for one moment are we suggesting you use a slide rule - we just like the picture.

Managers need information to understand trends in data and what drives them. It is essential that you question what the numbers mean. Only by doing this will you begin to know what is happening. Raw data has to be processed to give meaning and provide insights and information to manage marketing activities.


In my example here I pulled in data to a spreadsheet on world exports in clothing for 13 selected countries over 10 years. My purpose was to look for patterns that revealed trends in the data. If for example, you were to examine exports from China there is a doubling in exports but there is one year in the period were exports fell and you would want to know why that happened given the trend. Also, if you were to examine other data such as labour cost in the sector in these selected countries you would find that they were low relative to Europe and the USA; these being the markets where  most of these clothes are exported.


In my second chart example you can see where the UK imports the majority of its textiles and clothing from. 


Creating visualisations of data makes it easier for us to spot trends in data. We can summarise marketing data in a variety of ways. If we want to predict future trends we have to combine data that can give us a clear picture. Forecast data are used to do just that using simple linear regression and correlation or we may need to run multiple regression forecasts. Simple forecasts may use sales over time. Whereas multiple regressions predict a dependent variable Y (e.g. Sales) by identifying a number of independent variables x1,x2,x3...xn (e.g. GNP per head of population, percentage of GNP spent on Y and so on) that cause changes in Y.

Prof Hines © 2017

Multiple sources of data are essential to avoid over reliance on a single source. If you really want to understand what is happening different data sources will help you know.

"The ability to collect more data, and the potential of models based on big data in all its forms are alluring, But as a way of forecasting, overconfidence in data may give false confidence." MRS  (2017)

MRS. (2017). Prediction and planning in an uncertain world London: MRS.

Prof Hines © 2017

Prof Hines has developed two workshops to provide professional marketers, business owners and others with skills to develop their own dashboards and marketing analytics to understand their customer data. Details of dates and how to book a place will appear here.


Each of the two days can be attended as part 1 and part 2 of the same programme or  you can attend each one separately. 


Workshop 1

Will take participants from little knowledge to be able to do simple but powerful analysis using nothing more than aritmetic in the main. So if you are fearful of numbers there is no need to be.


Workshop 2 builds on the skills and knowledge from day 1 to take participants to the next level.